Вы находитесь на странице: 1из 346

DEFOAMING

SURFACTANT SCIENCE SERIES

FOUNDING EDITOR

MARTIN J. SCHICK
1918-1998

SERIES EDITOR

ARTHUR T. HUBBARD
Santa Barbara Science Project
Santa Barbara, California

ADVISORY BOARD

DANIEL BLANKSCHTEIN ERIC W. KALER


Department of Chemical Department of Chemical
Engineering Engineering
Massachusetts Institute of University of Delaware
Technology Newark, Delaware
Cambridge, Massachusetts
CLARENCE MILLER
S. KARABORNI Department of Chemical
Shell International Petrolewn Engineering
Company Limited Rice University
London, England Houston, Texas

LISA B. QUENCER DONRUBINGH


The Dow Chemical Company The Procter & Gamble Company
Midland, Michigan Cincinnati, Ohio

JOHN F. SCAMEHORN BERENDSMIT


Institute for Applied Surfactant Shell International Oil Products
Research B.V.
University of Oklahoma Amsterdam, The Netherlands
Norman, Oklahoma
JOHN TEXTER
P. SOMASUNDARAN Strider Research Corporation
Henry Krumb School of Mines Rochester, New York
Colwnbia University
New York, New York
1. Nonionic Surfactants, edited by Martin J. Schick (see also
Volumes 19, 23, and 60)
2. Solvent Properties of Surfactant Solutions, edited by
Kozo Shinoda (see Volume 55)
3. Surfactant Biodegradation, R. D. Swisher (see Volume 18)
4. Cationic Surfactants, edited by Eric Jungermann (see also
Volumes 34, 37, and 53)
5. Detergency: Theory and Test Methods (in three parts), edited by
W G. Cutler and R. C. Davis (see also Volume 20)
6. Emulsions and Emulsion Technology (in three parts), edited by
Kenneth J. Lissant
7. Anionic Surfactants (in two parts), edited by Warner M. Linfield
(see Volume 56)
8. Anionic Surfactants: Chemical Analysis, edited by John Cross
9. Stabilization of Colloidal Dispersions by Polymer Adsorption,
Tatsuo Sato and Richard Ruch
10. Anionic Surfactants: Biochemistry, Toxicology, Dermatology,
edited by Christian Gloxhuber (see Volume 43)
11. Anionic Surfactants: Physical Chemistry of Surfactant Action,
edited by E. H. Lucassen-Reynders
12. Amphoteric Surfactants, edited by B. R. Bluestein and
Clifford L. Hilton (see Volume 59)
13. Demulsification: Industrial Applications, Kenneth J. Lissant
14. Surfactants in Textile Processing, Arved Datyner
15. Electrical Phenomena at Interfaces: Fundamentals,
Measurements, and Applications, edited by Ayao Kitahara
and Akira Watanabe
16. Surfactants in Cosmetics, edited by Martin M. Rieger
(see Volume 68)
17. lnterfacial Phenomena: Equilibrium and Dynamic Effects,
Clarence A. Miller and P. Neogi
18. Surfactant Biodegradation: Second Edition, Revised
and Expanded, R. D. Swisher
19. Nonionic Surfactants: Chemical Analysis, edited by John Cross
20. Detergency: Theory and Technology, edited by W Gale Cutler
and Erik Kissa
21. lnterfacial Phenomena in Apolar Media, edited by
Hans-Friedrich Eicke and Geoffrey D. Parfitt
22. Surfactant Solutions: New Methods of Investigation, edited by
RaoulZana
23. Nonionic Surfactants: Physical Chemistry, edited by
Martin J. Schick
24. Microemulsion Systems, edited by Henri L. Rosano
and Marc Clausse
25. Biosurfactants and Biotechnology, edited by Naim Kosaric,
W. L. Cairns, and Neil C. C. Gray
26. Surfactants in Emerging Technologies, edited by Milton J. Rosen
27. Reagents in Mineral Technology, edited by P. Somasundaran
and Brij M. Moudgil
28. Surfactants in Chemical/Process Engineering, edited by
Darsh T. Wasan, Martin E. Ginn, and Dinesh 0. Shah
29. Thin Liquid Films, edited by I. B. Ivanov
30. Microemulsions and Related Systems: Formulation, Solvency,
and Physical Properties, edited by Maurice Bourrel
and Robert S. Schechter
31. Crystallization and Polymorphism of Fats and Fatty Acids,
edited by Nissim Garti and Kiyotaka Sato
32. lnterfacial Phenomena in Coal Technology, edited by
Gregory D. Botsaris and Yuli M. Glazman
33. Surfactant-Based Separation Processes, edited by
John F. Scamehorn and Jeffrey H. Harwell
34. Cationic Surfactants: Organic Chemistry, edited by
James M. Richmond
35. Alkylene Oxides and Their Polymers, F. E. Bailey, Jr.,
and Joseph \I. Koleske
36. lnterfacial Phenomena in Petroleum Recovery, edited by
Norman R. Morrow
37. Cationic Surfactants: Physical Chemistry, edited by
Donn N. Rubingh
and Paul M. Holland
38. Kinetics and Catalysis in Microheterogeneous Systems, edited
by M. Gratzel and K. Kalyanasundaram
39. lnterfacial Phenomena in Biological Systems, edited by
Max Bender
40. Analysis of Surfactants, Thomas M. Schmitt (see Volume 96)
41. Light Scattering by Liquid Surfaces and Complementary
Techniques, edited by Dominique Langevin
42. Polymeric Surfactants, lrja Piirma
43. Anionic Surfactants: Biochemistry, Toxicology, Dermatology.
Second Edition, Revised and Expanded, edited by
Christian Gloxhuber and Klaus Kunst/er
44. Organized Solutions: Surfactants in Science and Technology,
edited by Stig E. Friberg and Bjorn Lindman
45. Defoaming: Theory and Industrial Applications, edited by
P. R. Garrett
46. Mixed Surfactant Systems, edited by Keizo Ogino
and Masahiko Abe
47. Coagulation and Flocculation: Theory and Applications,
edited by Bohuslav Dobias
48. Biosurfactants: Production Properties Applications, edited by
Nairn Kosaric
49. Wettability, edited by John C. Berg
50. Fluorinated Surfactants: Synthesis Properties Applications,
Erik Kissa
51. Surface and Colloid Chemistry in Advanced Ceramics
Processing, edited by Robert J. Pugh and Lennart Bergstrom
52. Technological Applications of Dispersions, edited by
Robert B. McKay
53. Cationic Surfactants: Analytical and Biological Evaluation, edited
by John Cross and Edward J. Singer
54. Surfactants in Agrochemicals, Tharwat F. Tadros
55. Solubilization in Surfactant Aggregates, edited by
Sherril D. Christian and John F. Scamehorn
56. Anionic Surfactants: Organic Chemistry, edited by
Helmut W. Stache
57. Foams: Theory, Measurements, and Applications, edited by
Robert K. Prud'homme and Saad A. Khan
58. The Preparation of Dispersions in Liquids, H. N. Stein
59. Amphoteric Surfactants: Second Edition, edited by
Eric G. Lomax
60. Nonionic Surfactants: Polyoxyalkylene Block Copolymers, edited
by Vaughn M. Nace
61. Emulsions and Emulsion Stability, edited by Johan Sjoblom
62. Vesicles, edited by Morton Rosoff
63. Applied Surface Thermodynamics, edited by A. W. Neumann
and Jan K. Spelt
64. Surfactants in Solution, edited by Arun K. Chattopadhyay
and K. L. Mittal
65. Detergents in the Environment, edited by Milan Johann
Schwuger
66. Industrial Applications of Microemulsions, edited by Conxita
So/ans and Hironobu Kunieda
67. Liquid Detergents, edited by Kuo-Yann Lai
68. Surfactants in Cosmetics: Second Edition, Revised and
Expanded, edited by Martin M. Rieger and Linda D. Rhein
69. Enzymes in Detergency, edited by Jan H. van Ee, Onno Misset,
and Erik J. Baas
70. Structure-Performance Relationships in Surfactants, edited by
Kunio Esumi and Minoru Ueno
71. Powdered Detergents, edited by Michael S. Showell
72. Nonionic Surfactants: Organic Chemistry, edited by
Nico M. van Os
73. Anionic Surfactants: Analytical Chemistry, Second Edition,
Revised and Expanded, edited by John Cross
74. Novel Surfactants: Preparation, Applications, and
Biodegradability, edited by Krister Holmberg
75. Biopolymers at Interfaces, edited by Martin Malmsten
76. Electrical Phenomena at Interfaces: Fundamentals,
Measurements, and Applications, Second Edition, Revised and
Expanded, edited by Hiroyuki Ohshima and Kunio Furusawa
77. Polymer-Surfactant Systems, edited by Jan C. T. Kwak
78. Surfaces of Nanoparticles and Porous Materials, edited by
James A. Schwarz and Cristian I. Contescu
79. Surface Chemistry and Electrochemistry of Membranes, edited
by Torben Smith Serensen
80. lnterfacial Phenomena in Chromatography, edited by
Emile Pefferkorn
81. Solid-Liquid Dispersions, Bohuslav Dobias, Xueping Qiu,
and Wolfgang von Rybinski
82. Handbook of Detergents, editor in chief: Uri Zoller
Part A: Properties, edited by Guy Broze
83. Modern Characterization Methods of Surfactant Systems,
edited by Bernard P. Binks
84. Dispersions: Characterization, Testing, and Measurement,
Erik Kissa
85. lnterfacial Forces and Fields: Theory and Applications,
edited by Jyh-Ping Hsu
86. Silicone Surfactants, edited by Randal M. Hill
87. Surface Characterization Methods: Principles, Techniques, and
Applications, edited by Andrew J. Milling
88. lnterfacial Dynamics, edited by Nikola Ka/lay
89. Computational Methods in Surface and Colloid Science, edited
by Malgorzata Bor6wko
90. Adsorption on Silica Surfaces, edited by Eugene Papirer
91. Nonionic Surfactants: Alkyl Polyglucosides, edited by Dieter
Balzer and Harald Liiders
92. Fine Particles: Synthesis, Characterization, and Mechanisms of
Growth, edited by Tadao Sugimoto
93. Thermal Behavior of Dispersed Systems, edited by Nissim Garti
94. Surface Characteristics of Fibers and Textiles, edited by
Christopher M. Pasto· .md Paul Kiekens
95. Liquid Interfaces in Chemical, Biological, and Pharmaceutical
Applications, edited by Alexander G. Volkov
96. Analysis of Surfactants: Second Edition, Revised and Expanded,
Thomas M. Schmitt
97. Fluorinated Surfactants and Repellents: Second Edition, Revised
and Expanded, Erik Kissa
98. Detergency of Specialty Surfactants, edited by Floyd E. Friedli
99. Physical Chemistry of Polyelectrolytes, edited by Tsetska
Radeva
100. Reactions and Synthesis in Surfactant Systems, edited by
John Texter
101. Protein-Based Surfactants: Synthesis, Physicochemical
Properties, and Applications, edited by lfendu A. Nnanna and
Jiding Xia
102. Chemical Properties of Material Surfaces, Marek Kosmulski
103. Oxide Surfaces, edited by James A. Wingrave
104. Polymers in Particulate Systems: Properties and Applications,
edited by Vincent A. Hackley, P. Somasundaran,
and Jennifer A. Lewis
105. Colloid and Surface Properties of Clays and Related Minerals,
Rossman F. Giese and Carel J. van Oss
106. lnterfacial Electrokinetics and Electrophoresis, edited by
Angel II. Delgado
107. Adsorption: Theory, Modeling, and Analysis, edited by
J6zsefT6th
108. lnterfacial Applications in Environmental Engineering, edited by
Mark A. Keane
109. Adsorption and Aggregation of Surfactants in Solution, edited by
K. L. Mittal and Dinesh 0. Shah
110. Biopolymers at Interfaces: Second Edition, Revised and
Expanded, edited by Martin Malmsten
111. Biomolecular Films: Design, Function, and Applications, edited
by James F. Rusting
112. Structure-Performance Relationships in Surfactants: Second
Edition, Revised and Expanded, edited by Kunio Esumi
and Minoru Ueno
113. Liquid lnterfacial Systems: Oscillations and Instability, Rudolph
II. Birikh, Vladimir A. Briskman, Manuel G. Velarde,
and Jean-Claude Legros
114. Novel Surfactants: Preparation, Applications, and
Biodegradability: Second Edition, Revised and Expanded,
edited by Krister Holmberg
115. Colloidal Polymers: Synthesis and Characterization, edited by
Abdelhamid Elaissari
116. Colloidal Biomolecules, Biomaterials, and Biomedical
Applications, edited by Abdelhamid Elaissari
117. Gemini Surfactants: Synthesis, lnterfacial and Solution-Phase
Behavior, and Applications, edited by Raoul Zana and Jiding Xia
118. Colloidal Science of Flotation, Anh II. Nguyen
and Hans Joachim Schulze
119. Surface and lnterfacial Tension: Measurement, Theory,
and Applications, edited by Stanley Hart/and
120. Microporous Media: Synthesis, Properties, and Modeling,
Freddy Romm
121. Handbook of Detergents, editor in chief: Uri Zoller
Part B: Environmental Impact, edited by Uri Zoller
122. Luminous Chemical Vapor Deposition and Interface Engineering,
Hirotsugu Yasuda
123. Handbook of Detergents, editor in chief: Uri Zoller
Part C: Analysis, edited by Heinrich Waldhoff
and Rudiger Spilker
124. Mixed Surfactant Systems: Second Edition, Revised and
Expanded, edited by Masahiko Abe and John F. Scamehorn
125. Dynamics of Surfactant Self-Assemblies: Micelles,
Microemulsions, Vesicles and Lyotropic Phases, edited by
RaoulZana
126. Coagulation and Flocculation: Second Edition, edited by
Hansjoachim Stechemesser and Bohulav Dobias
DEFOAMING
Theory and
Industrial Applications

Edited by
P. R. Garrett

~ Taylor&Francis
~ Taylor&FrancisGroup
Boca Raton London New York Singapore

A CRC title, part of the Taylor & Francis imprint, a member of the
Taylor & Francis Group, the academic division of T&F lnforma pie.
Published in 1992 by
CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 1992 by Taylor & Francis Group, LLC


CRC Press is an imprint of Taylor & Francis Group
No claim to original U.S. Government works

15 14 13 12 11 10 9 8 7
International Standard Book Number 0-8247-8770-6 (Hardcover)
This book contains information obtained from authentic and highly regarded sources. Reprinted material is
quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts
have been made to publish reliable data and information, but the author and the publisher cannot assume
responsibility for the validity of all materials or for the consequences of their use.
No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic,
mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission from the publishers.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only
for identification and explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Catalog record is available from the Library of Congress

lldl informa I
Taylor & Francis Group
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at


is the Academic Division of T&F lnforma pie. http://www.crcpress.com
Preface

Problems-associated with excessive foaming occur in a surprisingly wide


range of situations. Development of suitable antifoams has therefore been a
necessity for the successful solution of foaming problems in industries as
different as textile dyeing and detergent manufacture. Arguably, the problem
of controlling excessive foam is generally of equal importance to that of
generating stable foams. Yet this problem usually receives scant attention
in the few textbooks that deal with the subject of foams. It seems timely,
therefore, to redress the balance and devote a text exclusively to the subject
of defoaming.
In many areas of surface and colloid science, the sheer complexity of
phenomena often means that technology has actually led scientific under-
standing. The development of antifoams involving the interaction of one
disperse phase (gas) with at least one other disperse phase (emulsion droplets
or particles) is perhaps such an area. By including a chapter in this volume
concerning the current theoretical understanding of the mode of action of
antifoams, it is hoped that some of the association of the subject with the
mysterious arts will be dispelled.
In many cases, the selection of an antifoam that will adequately control
the foam is only the beginning of a solution to a foaming problem. Attendant
considerations include the amount of antifoam required, the state of disper-
sion of the antifoam, and the possibility of the antifoam interacting adversely
with some aspect of a process. These considerations may often be of ov-
erriding importance and may therefore affect the selection of the antifoam.
They are also usually specific to particular applications. For this reason the
Iii
Iv Preface

volume is divided into chapters concerning different industrial applications


of antifoams so that detailed descriptions of the problems and solutions spe-
cific to given industries may be dealt with. Not all such applications are
covered in this one volume, but a sufficiently wide range is included to give
some indication of the appropriate approaches for any foam control problem.
Also, a measure of coherence is maintained throughout the book despite the
varied nature of the industries considered so that nomenclature and defini-
tions of terms are consistent.
This book would not have been possible without the timely contributions
of the various authors. Their patience in enduring the many editorial changes,
which a volume such as this entails, is gratefully acknowledged. Finally,
gratitude is due to Unilever Research for the provision of secretarial assis-
tance which greatly facilitated the preparation of this volume.

P. R. Garrett
Contents

Preface iii
Contributors vii
1 The Mode of Action of Antifoams 1
P. R. Garrett
2 Antifoams for Nonaqueous Systems in the Oil Industry 119
Ian C. Callaghan
3 Defoaming in the Pulp and Paper Industry 1S1
S. Lee Allen, Lawrence H. Allen, and Ted H. Flaherty
4 Application of Antifoams in Pharmaceuticals 177
Rolland Berger
S High-Performance Antifoams for the Textile Dyeing Industry 193
George C. Sawicki
6 Foam Control in Detergent Products 221
Horst Ferch and Wolfgang Leonhardt
7 Antifoams for Paints 269
Maurice R. Porter
8 Surfactant Antifoams 299
Trevor G.-Blease, J. G. Evans, L. Hughes, and Philippe Loll

Index 325
v
Contributors

Lawrence H. Allen Pulp and Paper Research Institute of Canada, Point


Claire, Quebec, Canada

S. Lee Allen Pulp and Paper Research Centre, McGill University, Mon-
treal, Quebec, Canada

Rolland Berger Th. Goldschmidt AG, Chemische Fabriken, Essen,


Germany

Trevor G. Blease Research and Technology Department, ICI Surfactants,


Wilton, Middlesborough, United Kingdom

Ian C. Callaghan Colloid Science Branch, BP Research, Sunbury-on-


Thames, Middlesex, United Kingdom

J. G. Evans Research and Technology Department, ICI Surfactants, Wil-


ton, Middlesborough, United Kingdom

Horst Ferch* Applied Technologies Silicas and Silicates, Degussa AG,


Hanau, Germany

*Retired.

vii
viii Contributors

Ted H. Flaherty Dorset Industrial Chemicals, Ltd., Chateauguay, Que-


bec, Canada

P. R. Garrett Port Sunlight Laboratory, Uniliver Research, Bebington,


Wirral, United Kingdom

L. Hughes Research and Technology Department, ICI Surfactants, Wil-


ton, Middlesborough, United Kingdom

Wolfgang Leonhardt Applied Technologies Silicas and Silicates, De-


gussa AG, Hanau, Germany

Philippe Loll ICI Surfactants, Everberg, Belgium

Maurice R. Porter Maurice R. Porter and Associates, Sully, South Gla-


morgan, United Kingdom

George C. Sawicki Department of Science and Technology, Dow Corning


Europe, Brussels, Belgium
1
The Mode of Action of Antifoams

P. R. GARRETT Unilever Research, Bebington, Wirral, United


Kingdom

I. Introduction 2
II. The Stability of Foams 3
A. Surface tension gradients and foam film stability 4
B. Disjoining forces and foam film stability 8
III. Antifoams and Surface Activity 11
A. Antifoam effects in homogeneous systems 11
B. Duplex oil films, oil lenses, and antifoam mechanism 13
C. Surface tension gradients and theories of antifoam
mechanism in heterogeneous systems 19
IV. Inert Hydrophobic Particles and Capillary Theories of
Antifoam Mechanism for Aqueous Systems 30
A. Early work 30
B. Contact angles and particle bridging mechanism 33
C. Particle geometry and contact angle conditions for
foam film collapse 41
D. Particle size and kinetics 53
E. Capture of particles by bubbles 58
F. Melting of hydrophobic particles and antifoam behavior 63
V. Mixtures of Hydrophobic Particles and Oils as Antifoams
for Aqueous Systems 66
A. Antifoam synergy 66
B. Spreading behavior of antifoam oils on water 82
C. Role of the oil in synergistic oil/particle antifoams 87
D. Hypotheses concerning the role of the particles in
synergistic oil/particle antifoams 94
1
2 Garrett

E. Capture of oil droplets by bubbles and the


role of the particles 98
F. Antifoam dimensions and kinetics 11 l
Acknowledgments 113
References 113

I. INTRODUCTION
This chapter attempts a complete review of the various mechanisms pro-
posed for the action of antifoams over the past half-century. It is a feature
of this subject that these mechanisms, although plausible, are often specu-
lative. Thus, unequivocal experimental evidence is often lacking. Indeed,
the full theoretical implications of proposed mechanisms are also often not
fully developed. In the main, all of this derives from the extreme complexity
of the relevant phenomena. Foam is itself extremely complex, consisting of
(usually) polydisperse gas bubbles separated by draining films. These films
exhibit complicated hydrodynamics involving the distinct rheology of air-
liquid surfaces and, for thin films, colloidal interaction forces. The nature
of the foam film collapse processes which are intrinsic to a foam are still
imperfectly understood.
Antifoams are usually hydrophobic, finely divided, insoluble materials.
Their presence therefore further complicates the complexities associated with
foam. Indeed commercial antifoams for aqueous solutions usually consist of
hydrophobic particles dispersed in hydrophobic oils. The action of such
antifoams concerns the effect of a dispersion (of antifoam in foaming liquid)
of a dispersion (the antifoam) on yet a third dispersion (the foam).
Theories of antifoam mechanism appear to fall into two broad categories:
those which require the antifoam to be surface active at the air-liquid sur-
faces of the foaming liquid, and those which do not. Theories which require
the antifoam to be surface active associate antifoam behavior with an effect
on surface tension in the foam films which leads to film rupture. Theories
which do not require the antifoam to be surface active usually concern the
hydrophobic nature of the antifoam. They suppose that dewetting of anti-
foam entities in foam films produces capillary instabilities which lead to film
rupture.
The chapter is divided into four sections. The first concerns an outline
of the main processes which are believed to contribute to the stability of
foam films in the absence of antifoam. The second examines the relationship
between antifoam effects and surface activity of the antifoam. The third
section concerns inert hydrophobic particles and capillary theories of anti-
foam action in aqueous solutions. The last section concerns the mode of
Mode of Action of Antifoams 3

air

Polyederschaum

Kugelschaum

liquid

FIG. 1 Foam structure.

action of the mixtures of hydrophobic particles and oils which form the basis
of many of the commercial antifoam concoctions proposed for aqueous foams.

II. THE STABILITY OF FOAMS


Before considering the mode of action of antifoams, we review the factors
which contribute to the stability of a foam. A brief summary only is given
here. For more complete accounts the reader is referred to the many reviews
on the subject [1-5]. Our summary closely follows the excellent review by
Lucassen [2].
The structure of a typical foam formed by, say, shaking a surfactant so-
lution in a cylindrical vessel is shown schematically in Fig. I. In the lower
part of the foam, bubbles are spherical (so-called kugelschaum) and of small
size with a relatively low gas volume fraction. As the liquid drains out
of the foam, the bubbles distort to form polyhedra. This polyhedral foam
(polyederschaum) consists of plane-parallel films joined by channels called
plateau borders. The gas volume fraction is here relatively high and the
density low so that polyhedra first form at the top of the foam column.
4 Garrett

Throughout the foam there are differences in the sizes of adjacent bub-
bles. This will mean that differences in capillary pressure will exist between
the adjacent bubbles so that gas will diffuse from small to large bubbles.
The more soluble the gas in the continuous phase or the higher the partial
vapor pressure of the continuous phase the faster this process of bubble dis-
proportionation will proceed. There is no arrangement of bubbles in a foam
which permits elimination of this process. Thus, for example, if there is free
headspace above the foam the upper surface will consist of films of curved
section. This will mean diffusion of gas out of the upper layer of bubbles
into the headspace because of the capillary pressure implied by the curved
surface.
As the films at the top of the foam thin, they become more susceptible
to rupture by mechanical shock or vibration. Moreover, with some foams
rupture of films at a certain thickness is spontaneous. Films at the top of
the foam then tend to break first, and the foam collapses from the top down-
ward in a catastrophic cascade.
A. Surface Tension Gradients and Foam Film
Stability
Films formed by adjacent bubbles in a pure liquid are extremely unstable.
Pure liquids therefore do not form foams. This arises in part because of the
response of the films to any external force such as gravity. Consider, for
example, a vertical plane-parallel film in a gravity field. There is no reason
why any element of that film should move in response to the applied grav-
itational force with a velocity different from that of any adjacent element.
No velocity gradients in a direction perpendicular to the plane of the film
surface against the air will therefore exist. There will then be no viscous
shear forces opposing the effect of gravity. The film will exhibit plug flow
(resisted only by extensional viscous forces) with elements accelerated
downward tearing it apart. The process is depicted in Fig. 2a.
This behavior can be drastically altered if we arrange for a tangential
force to act in the plane of the liquid-air surface so that the surface is es-
sentially rigid. In the case of a vertical plane-parallel film of a viscous liquid
with such rigid surfaces, subject to gravity, a parabolic velocity profile will
develop as shown in Fig. 2b. This means that velocity gradients will exist
in a direction perpendicular to the film surfaces. A viscous stress will there-
fore be exerted at the air-liquid surface. This stress must be balanced by the
tangential force acting in the plane of the surface. That force can only be a
gradient of surface tension. This balance of viscous forces and surface ten-
sion gradients at the liquid-air surface can be written as
d"'f
---TJF -
AF _ (dUy) (1)
dy dx x=O
5

TTT
Mode of Action of Antifoams

(a)

(b)

FIG. 2 Velocity profiles in draining foam films: (a) plug flow; (b) flow with par-
abolic velocity profile when film surfaces are immobile.

where 'YAF is the air-liquid surface tension of the foaming liquid, TJF is the
viscosity, uy is the velocity of flow in the y direction, y is the vertical dis-
tance, and x is the horizontal distance in the film.
Thus, we find that if the force of gravity (or indeed any other force such
as that due to the capillary pressure caused by the curved plateau borders)
is to be resisted by the film, then a surface tension gradient must exist at
the air-liquid surface. In the case of a vertical film in the gravity field the
gradient is [2]
d"{AF pgh
(2)
dy 2
where h is the film thickness, p is the liquid density, and g is the acceleration
due to gravity. This gradient can only exist where differences of surface
composition can occur. We therefore require the presence of more than one
6 Garrett

component in the film. Indeed, it is possible to speculate that in the case


of, say, aqueous foams diffusion of water through the gas phase may rapidly
remove any differences in concentration between different parts of a foam
film if only one solute is present. In this case at least two solutes (or three
components) would be required.
Surface tension gradients due to differences in the surface excess of sol-
uble surface-active components may exist only when either the surface is
not in equilibrium with the bulk composition or there are concomitant dif-
ferences in bulk composition parallel to the surface. In the case of the former
the magnitudes of the gradients are of course determined by the rate of trans-
pott of surfactant to the relevant surfaces. With concentrated surfactant so-
lutions transport rates by diffusion will be rapid and surface tension gradients
will tend to be eliminated. Thus, it has occasionally been reported that foam-
abilities decline at extremely high concentrations of surfactant in aqueous
solution. Conversely, however, if foam films are denuded of surfactant be-
cause of extremely slow transport rates, then the maximum surface tension
gradients which can be achieved will be small. Such films will therefore be
susceptible to rupture when exposed to external stress. However, the com-
plex problem of assessing both the effect of rate of transport on the surface
tension gradients in foam films and the overall resultant impact upon foam
film stability, when subject to an external stress, has not apparently been
fully addressed.
Differences in bulk composition are possible in a thin foam film as a
result of stretching the film. If the film is sufficiently thin, then any stretch-
ing causes a depletion of the bulk phase surfactant solution between the air-
liquid surfaces of the foam film as more surfactant adsorbs on those surfaces.
Distances perpendicular to the film are small so that, provided the stretching
occurs reasonably slowly, equilibrium inside the film element may be al-
ways maintained. Depletion of bulk phase surfactant concentration will
therefore necessarily mean an increase of the surface tension of the film as
it is stretched. This will, however, only occur if reduction of surfactant con-
centration causes a concomitant increase in surface tension. In the case of
a pure surfactant at concentrations above the critical micelle concentration
(cmc), this may not always happen.
We find then that it is possible to generate a surface tension gradient in
a foam film by stretching various elements of the film to different extents.
The increase in surface tension due to stretching imparts an elasticity to the
film. This property of foam films was first recognized by Gibbs [6] and is
usually referred to as the Gibbs elasticity EG. It is defined as

2dyAF 2dyAF
EG=--= - - - (3)
d In A d In h
Mode of Action of Antifoams 7

50

h = 10-- 4 cm
40

30
cmc
rf\

20
h=2x10- 4 cm

10

5 10

concentration SDS I (M x 103)

FIG. 3 Gibbs elasticities of submicellar sodium dodecylsulfate solutions. (From


Ref. 2.)

where A is the film area and the factor 2 arises because of the two surfaces.
A plot of ea against concentration for a submicellar aqueous solution of
sodium dodecylsulfate (SDS) is shown in Fig. 3 by way of example. Here
we see that, except at very low concentrations, decreases in film thickness
at constant concentration produce increases in Gibbs elasticity so that (iJeo/
i!h)c :s 0. Thus, as the film becomes thinner stretching will cause a relatively
greater depletion of surfactant in the intralamellar liquid and the surface
tension will rise to a greater extent.
The plot of Gibbs elasticity against concentration shown in Fig. 3 clearly
reveals a maximum at concentration cmax· At extremely low concentrations
of surfactant we find that upon stretching of the film there is essentially no
contribution from the intralamellar liquid, and the surfactant behaves as an
insoluble monolayer. Here with increase in surfactant concentration both the
surface excess and the elasticity of the monolayer increase. However, further
increases in the surfactant concentration will eventually mean that it signif-
icantly exceeds that required to compensate for stretching of the air-liquid
surface, so ea """' 0. These two opposing consequences of increasing con-
centration conspire to produce the maximum in a plot of Gibbs elasticity.
8 Garrett

Lucassen [2] considers the effect of Gibbs elasticity on the development


of stabilizing surface tension gradients in a foam film in the gravity field.
He points out that if E 0 decreases as the film is stretched it will tend to be
dynamically unstable. Under these circumstances any stretching force will
tend to increase the area of the thinnest part of the film. Such situations will
tend to prevail for films formed at concentrations on the low side of Cmax.
Thus, we can write

dEo = (aEo) + (aEo) de


(4)
dh ah c ac h dh
where c is the concentration. We can therefore have dE 0 /dh > 0 if c < Cmax•
because then (aE 0 /ac)h > 0, dc/dh > 0, and (aEc/ah)c ~ 0 as c ~ 0.
For vertical films prepared from concentrated solutions the requirement
that the surface tension gradient satisfy Eq. 2 implies that rapid stretching
will occur. In the case of micellar solutions of certain pure surfactants this
may require achievement of submicellar concentrations in order that Ea >
0 and therefore d-yAFf dy > 0. For vertical films prepared from extremely
dilute surfactant solutions where c < cm.., Lucassen [2] shows that the mag-
nitude of Eo for thin elements at the top of the film may be less than that
of thinner elements lower down. Any force acting on the film, such as an
increase in weight as it grows, could mean catastrophic extension of the
thinnest elements because of their lower Gibbs elasticities.
In summary, then, we find that surface tension gradients are necessary if
freshly formed foam films are to survive. These gradients may occur if sur-
face tensions depart from equilibrium values. This will happen when foam
film air-liquid surfaces are expanded at rates which are fast so that equilib-
rium with the bulk surfactant concentration cannot be maintained. They may
also occur when films are thin so that stretching may deplete intralamellar
bulk phase to give rise to a Gibbs elasticity. Unfortunately there are few
experimental observations which clearly reveal the importance of surface
tension gradients in determining foam behavior. Perhaps the best examples
are reported by Malysa et al. [7) and Prins [8].

B. Disjoining Forces and Foam Film Stability


Even though foam films may be stabilized by effective surface tension gra-
dients they may still succumb to rupture due to the action of van der Waals
forces. These have a significant effect on foam film stability at film thick-
nesses where the intralamellar molecules cease to have the properties of
matter in bulk because of the perturbing influence of the two contiguous air-
liquid surfaces. In effect, molecules in a thin film have a higher chemical
potential (or vapor pressure) than matter in bulk at the same external pressure
Mode of Action of Antifoams

liquid movement due to

--------+- --- ----t----


- disjoining force

\
unperturbed
film surfaces

--t- --- ---+--- I


liquid movement due to
capillary force _.

FIG. 4 Sinusoidal thickness perturbations in a thin liquid foam film.

because of the weaker attractive force field due to the limited number of
molecules present. Such molecules will therefore tend to spontaneously transfer
to any adjacent bulk phase. This process is said to be driven by a "disjoining
pressure" n which is negative when the film shows a tendency to become
thinner. Clearly as the film becomes thinner the disjoining pressure becomes
more negative so that
dfl
->0 (5)
dh
A perturbation of the film thickness which produces thin and thick regions
will tend to grow spontaneously because molecules in the thin region will
transfer to the thick regions. However, any perturbation of a film to produce
thick and thin regions must also inevitably increase the surface area of the
film. This will increase the number of molecules in the relatively weak at-
tractive force field close to the air-liquid surface. Such an increase in surface
area will therefore be resisted by an opposing force-the surface tension.
Thickness perturbations, of a thermal or mechanical nature, may be con-
sidered to be of a wavelike nature. A symmetrical sinusoidal perturbation
is shown in Fig. 4. Here the thin part of the film is subject to two opposing
forces. Thus, a capillary pressure due to the surface tension tends to suck
liquid back into the thin part of the film, and a disjoining force tends to
push liquid away.
The magnitude of the capillary pressure is determined by the curvature
of the film surface. For a given amplitude of the perturbation the curvature
10 Garrett

is determined by the wavelength. Thus, the shorter the wavelength the more
marked the curvature and the stronger the capillary pressure. Vrij and
Overbeek [9] deduce a critical wavelength Acri• above which disjoining forces
will dominate over the capillary pressure and the perturbation will sponta-

J
neously grow. The critical wavelength is
21T 2'Y AF 1/2
A. -
crit -
[ --
dll/dh
(6)

The rate of growth of the perturbation will increase with increasing wave-
length A. for A. > Acrit because the damping effect of the capillary pressure
will decrease. However, for sufficiently long wavelengths the rate of growth
will eventually begin to decline because of the increased distances over which
the film liquid has to be moved against viscous resistance. An optimum
wavelength of \/211.crit for the maximum rate of growth of a perturbation
therefore exists [9].
Film collapse is supposed to occur when the amplitude of the fastest growing
perturbation equals the thickness of the film. Vrij and Overbeek [9] have
produced the simplest description of this process. They calculate the mini-
mum total time for film drainage and subsequent growth of the fastest-grow-
ing perturbation. The average thickness of the film at the moment of rupture
is the critical thickness hcrit· Experimental measurements of hcrit for micro-
scopic foam films are in the region of a few tens of nanometers (see, for
example, [ IO]).
Clearly this description ignores any contribution to disjoining forces from
stabilizing interactions across a thin film due to adsorbed surfactant. These
could arise from overlapping double layers or steric factors due to interaction
of bulky surfactant head groups. Such forces could mean Il > 0 and dll/
dh < 0.
With many surfactants at sufficiently high concentrations (and therefore
adsorption levels) the growth of perturbations at a particular critical thick-
ness results not in foam film rupture but in the formation of spots of thin
metastable film. These spots then grow to cover the whole film area. This
film is thin so that destructive interference of reflected light occurs. It there-
fore appears black in reflected light.
The concentration at which this transition from rupture to black spot for-
mation occurs is cblack· The transition probably occurs because of the effects
of changes in the adsorption layer of surfactant on the thickness dependence
of Il and dH/dh. Foam film stability tends to increase markedly at concen-
trations above cblack·
Values for cblack in microscopic films have been measured for a number
of surfactants [ 11]. In the case of aqueous solutions of surfactants cblack is
generally significantly lower than the cmc. Thus, for example, for sodium
Mode of Action of Antifoams

dodecylsulfate, Cbtack = 1.6 X 10- 6 Mfl1] and the cmc = 8.4 X 10- 3 M
[12], and for dodecyl hexaethyleneglycol cbtack = 4.9 x 10- 6 M [11] and
the cmc = 8.7 x 10 5 M [13] at 25°C.
We have seen then that film rupture may occur because either surface
tension gradients are not sufficiently high to enable the film to withstand
stress or because dfl/dh is always positive so that rupture is inevitable at a
certain critical thickness. However, both phenomena are associated with low
concentrations of surfactant (at least if we consider films formed slowly so
that equilibrium between the air-liquid surface and the intralamellar liquid
is maintained). Thus, we have Cmax ~ cmc for dEG/dh < 0 and Chtack ~ cmc.
Elimination of both causes of rupture should therefore be readily achieved
at sufficiently high concentrations of surfactant. The poor discrimination in
foam behavior often found with relatively concentrated aqueous micellar so-
lutions of surfactants may well be attributable to that cause. Interesting dif-
ferences in foam behavior are, however, often revealed when antifoam is
added to the solution.

111. ANTIFOAMS AND SURFACE ACTIVITY


A. Antifoam Effects in Homogeneous Systems
One of the earliest generalizations concerning antifoams states that they must
be present as undissolved particles (or droplets) in the liquid to be defoamed
f 14-16]. Indeed the presence of antifoam materials at concentrations lower
than the solubility limit can even enhance foamability [ 17, 18]. One well-
known example concerns the foam-enhancing effect of dissolved polydi-
methylsiloxanes on hydrocarbon lube oils [ 17). However, the generalization
that antifoams must be present as undissolved entities has occasionally been
repeated with some equivocation [18-20). A number of authors in fact re-
port experimental results which purport to show antifoam effects due to ad-
ditives which are dissolved in the foaming solution [21-23]. Thus, Ross and
Haak [21], for example, identify two types of antifoam behavior associated
with the effect of materials like tributyl phosphate and methylisobutyl car-
binol on the foam behavior of aqueous solutions of surfactants such as so-
dium dodecylsulfate and sodium oleate. Wherever the oil concentration ex-
ceeds the solubility limit, emulsified droplets of oil contribute to an effective
antifoam action. However, if micelles of the surfactant are present, then the
oil is solubilized to a certain limiting concentration. It is claimed [21,24]
that a weak antifoam effect is associated with the presence of such solu-
bilized oils in the micelles. The consequences of all this behavior are re-
vealed if, for example, tributyl phosphate is added to micellar solutions of
sodium oleate [21] at concentrations below the solubilization limit. A marked
12 Garrett

decrease in foamability is found immediately after dispersing the oil. As the


oil becomes slowly solubilized, the foamability increases. However, even
after the oil is completely solubilized the foamability is still apparently less
than that intrinsic to the uncontaminated surfactant solution [21].
Ross and Haak [21] have also studied the effect of solubilized tributyl
phosphate and methylisobutyl carbinol on the dynamic surface tensions of
micellar solutions of sodium dodecyl sulfate and sodium oleate, using the
oscillating jet technique. They claim that the rate of decrease of surface
tension is significantly enhanced by the presence of these solubilisates, pre-
sumably because of their effect upon transport of micellar surfactant to the
air-water surface. Ross and Haak [21] used these findings in seeking an
explanation for the weak antifoam effect attributable to these solubilisates.
They suggest that a major factor in determining foam film stability is res-
toration of foam film thickness after stretching. Thus, if a thick foam film
is rapidly stretched, the surface tension will increase due to the finite time
taken for surfactant to adsorb onto the expanded surface. The difference in
surface tension between the stretched film element and the surrounding un-
perturbed film elements or plateau borders will lead to a Marangoni flow
which will tend to restore the film thickness.
If, however, the rate of surface tension relaxation is too rapid, then the
driving force for such restoration will disappear before it can occur to any
significant extent. This will mean that the film can extend without effective
resistance. According to Ross and Haak [21J the "resilience" is then lost.
It is therefore implied by these authors that extension of the film would then
proceed until rupture occurs. A similar argument had been advanced earlier
by Burcik [25].
There are some obvious difficulties with this hypothesis. Thus, it con-
cerns extremely thick films where only rate processes can give rise to surface
tension increases on film stretching. However, as a film thins, such stretch-
ing will give rise to surface tension increases as a result of depletion of
surfactant by adsorption (to give a finite Gibbs elasticity). This effect would
appear even when adsorption is instantaneous. Therefore complete loss of
resilience by increases in the rate of adsorption would be confined to ex-
tremely thick films-resilience would eventually be restored if the film was
sufficiently stretched. Another difficulty concerns the time scale over which
the supposed film stretching occurs. Thus, the dynamic surface tensions upon
which the hypothesis is based were all measured for surface ages < 3 x
10- 2 s (the range of the oscillating jet technique). However, suppose, for
example, that the stretching occurs more slowly than Ross and Haak [21]
imply so that the relevant time scale is, say, 10- 1 s and solutions both with
and without antifoam have reached equilibrium at that surface age. The ad-
verse effect of solubilized tributyl phosphate and methylisobutyl carbinol on
Mode of Action of Antifoams 13

foamability would then clearly not concern their effect on dynamic surface
tension over the irrelevant time scale probed by the oscillating jet method.
Absence of either independent evidence concerning the relevant time scale
or of dynamic surface tensions over wider time scales means that the hy-
pothesis of Ross and Haak [21] is inadequately tested. We should also note
that, although rapid transport of surfactant to air-water surfaces is not proven
to lead necessarily to low foamability or foam stability, it is easy to see that
extremely slow transport would. Thus, for example, if bubbles of gas ap-
proach one another during foam generation, then film collapse and coales-
cence will inevitably occur if stabilizing surfactant is essentially absent from
the relevant air-water surfaces due to slow transport (see Sec. II). Burcik
[25] recognized this difficulty and compromised by suggesting that high
foamability requires "moderate" rates of surface tension lowering.
We may conclude by noting that, in general, antifoam effects appear to
require that the antifoam be undissolved in the foaming medium. Solubili-
zation of the antifoam in micelles largely restores the foamability. However,
there is some evidence to suggest that the solubilized antifoams may have
a weak adverse effect on the foamability or foam stability of a micellar
solution [21]. We should note, however, that dissolved antifoam can even
enhance foamability [ 17, 18].

B. Duplex Oil Films, Oil lenses, and Antifoam


Mechanism
Antifoams are often largely composed of oils which are present as undis-
solved droplets in foaming solutions. Presence in the form of undissolved
droplets is not, however, a sufficient requirement. Thus, it has often been
proposed that such oils should also be surface active and exhibit the property
of spreading over the air-liquid surface of the foaming liquid [l,17,20,26,27].
The first proposal that spreading be associated with antifoam behavior was
made by Leviton and Leighton [28] more than half a century ago. Subse-
quently there have been numerous attempts to correlate the antifoam be-
havior of indissolved oils with their spreading behavior [20,22,29-31].
Here we are concerned with examining the possibility that duplex film
spreading by undissolved oils is a necessary aspect of their anti foam mech-
anism. Such spreading occurs when a thin film of oil with distinct upper
and lower surfaces, having the surface tensions of the bulk oil, spreads over
the surface of the substrate liquid. The driving force for this process arises
from the relevant surface tensions.
If an oil which is dispersed as droplets in the foaming liquid is to spread
at the air-liquid surface of that liquid it must first emerge into the surface.
For this to happen spontaneously the entry coefficient E must be positive.
14 Garrett

Here E is defined by
E = 'YAF + 'YoF - 'YAo (7)
where 'Y AF is the air-liquid surface tension of the foaming solution, 'YoF is
the antifoam oil-foaming solution interfacial tension, and 'Y Ao is the anti foam
oil-air surface tension. Selection of the appropriate values of the three sur-
face tensions should take account of the extent to which the antifoam and
foaming solution have been preequilibrated. Thus, we could follow Ross
[20) and define two other entry coefficients,

E' = 'YAF + 'YbF - 'Y~o (8)

and
(9)

where the primes refer to surface tension values for the mutually saturated
antifoam and foaming fluid: E is usually described as the initial entry coef-
ficient, E' is the semi-initial entry coefficient where the antifoam droplets
are saturated with respect to the foaming liquid and the foaming liquid-air
surface is uncontaminated with antifoam, and E" is the final entry coefficient
where both fluids are mutually saturated.
If the oil is also to spread at the surface of the foaming liquid to form a
duplex film, then the spreading coefficient S must also be positive. Here we
may follow Harkins [32) and define an initial spreading coefficient
S = 'YAF - 'YoF - 'YAO (10)

a semi-initial spreading coefficient


S' = 'YAF - 'YbF - 'Y~o (11)

and a final spreading coefficient


(12)

These three spreading coefficients are exactly analogous to the three entry
coefficients in Eqs. 7-9.
If we have E > 0 and S < 0, then placing a drop of the oil on the surface
of the foaming liquid will form an oil lens and not a duplex film. Harkins
[32] claims that the final spreading coefficient S" is always negative, so that
an oil will not spread on the surface of the foaming liquid after it has become
saturated with respect to that oil. This means that if S > 0 the resulting
duplex film is unstable and will disproportionate to form oil lenses in equi-
librium with antifoam-contaminated air-foaming liquid surfaces character-
ized by a surface tension 'Y~F· Although this type of behavior is often found
in practice, Rawlinson and Widom [33] argue that it is not always found.
Mode of Action of Antifoams 15

Thus, in disagreement with Harkins [32), they state that we should have S"
:5 0 and not S" < 0, so that S" = 0 is allowed. This means that stable duplex
films can sometimes be found between mutually saturated fluids at equilibrium.
An additional complication associated with the relationships in Eqs. 7-
12 concerns the effect of the finite time required for surfactant to transport
to the relevant surfaces during foam generation. Thus, the surface tension
of the foaming liquid may be higher than equilibrium even if the liquid is
saturated with respect to the antifoam. We may then find, for example, that
S" > 0 under dynamic conditions because ('Y~F)dynamic > ('Y~F)equilibrium·
Ross [20] has advanced a hypothesis concerning the rupture of foam films
by oil droplets for which E > 0 and S > 0. Such oil droplets will, as we
have seen, emerge into the air-foaming liquid surface and spread as a duplex
film. If this were to happen with an oil droplet simultaneously emerging into
both surfaces of a foam film, a duplex film of oil would simultaneously
spread over both surfaces of the foam film. This would squeeze out the
original liquid in the film to produce a region composed entirely of antifoam
oil where rupture would occur. Presumably this mechanism would cease to
function if the foaming liquid and antifoam become mutually saturated so
that we have S" < 0 (if dynamic effects can be ignored). The antifoam effect
would therefore decay with time. Further addition of antifoam oil would
also be without effect.
Ross [20] interpreted some early results of Robinson and Woods [19],
using this hypothesis. Those results concerned the effect of various undis-
solved oils on the foam behavior of both aqueous and nonaqueous solutions
of surfactant. The oils included alkyl phosphates, alcohols (including diols),
fatty acid esters, and polydimethylsiloxane. The solutions were of aerosol
OT (sodium diethylhexyl sulfosuccinate) in either ethylene glycol or tri-
ethanolamine and sodium alkylbenzene sulfonate in water.
Robinson and Woods [19] observed that for these systems, wherever E
< 0, no antifoam effect is found. This then represents clear evidence that
a positive value of the initial entry coefficient is necessary for antifoam ac-
tion. In addition, however, Ross [20] noted that of the 40 combinations of
oils and solutions with E > 0 which exhibited antifoam behavior, 36 had S
> 0 and 4 had S < 0. He claimed that this finding represented "definite"
agreement with the proposed rupture mechanism which requires E > 0 and
S > 0. That conclusion of course ignores the finding that antifoam effects
can occur with S < 0. A rigorous conclusion would therefore be that either
spreading (i.e., S > 0) is not a necessary aspect of the antifoam behavior
of these oils or that two different mechanisms may operate corresponding
to the two sets of conditions E > 0, S > 0 and E > 0, S < 0.
In a later paper Ross and Young [29] find other examples of systems for
which E > 0 and S < 0 where antifoam effects are observed. Moreover,
16 Garrett

Okasaki et al. [22] report that both petroleum ether and olive oil spread
rapidly over the surface of 10- 2 M sodium dodecylsulfate solution but do
not yield antifoam effects with the same solution. By contrast these authors
find that undissolved phenol does not spread on 10- 2 M sodium dodecyl-
sulfate solution but does function as an effective antifoam (albeit at the rather
high antifoam concentration used in this work).
In a study of the antifoam effect of normal alcohols, Kruglyakov and
Taube [31) failed to find a correlation between initial spreading coefficients
Sand antifoam effectiveness. Here for octanol (and higher alcohols) on aqueous
solutions of a variety of surfactants the initial spreading coefficient was found
to be negative despite optimal antifoam effect with that alcohol under the
conditions studied. Indeed, in another paper Kruglyakov [34) states that among
the "numerous systems investigated by Ross and other workers there are
many cases where efficient foam breaking was observed at a negative spreading
coefficient" (i.e., when S < 0).
Kruglyakov and Koretskaya [35) have extended this study of normal al-
cohols to include the effect of surfactant concentration. They find that the
optimum chain length of alcohol for maximum antifoam effect increases
with decreasing concentration of surfactant. This behavior is shown in Fig.
5 for solutions of a blend of ethoxylated nonyl phenols (in 0.1 M KCl). Here
the limiting concentration of antifoam for the foam to exhibit instability is
plotted against alcohol chain length for various concentrations of surfactant.
Kruglyakov and Koretskaya [35) argue that the antifoam effect of these al-
cohols is due to the formation of "asymmetrical films" where an oil lens is
present on one side of a foam film. The configuration is illustrated sche-
matically in Fig. 6. Such lenses will form provided E > 0 and S < 0. Indeed,
as we have seen, even if S > 0 the resulting duplex film may be unstable
so that ultimately the conditions E" > 0 and S" < 0 may prevail and lenses
may form.
Kruglyakov [34] argues that the aqueous asymmetrical film separating an
oil droplet from the air-water surface is metastable in the same sense as a
symmetrical air-water-air foam film provided E" > 0. However, he also
argues that the asymmetrical film will in general have a different stability
from that of the symmetrical foam film with the same solution. This is sup-
posed to arise in part because of differences in surface excess of surfactant
at the oil-water and air-water surfaces. If the oil is polar, then the adsorption
of the surfactant at the oil-water surface will be lower than at the air-water
surface. In tum this will mean diminished stabilizing repulsive forces across
the film due, for example, to overlapping electrostatic double layers if the
surfactant is ionic. The stability of the film will therefore be diminished
provided the entry coefficient is still positive. Foam film collapse is then
supposed to occur when the relatively unstable aqueous film separating an
Mode of Action of Antifoams 17

limiting alcohol
concentration
%
101

10°

10- 1

10- 2 6

.._~~-'-~~-'-~~-'-~~-'-·____J
4 6 8 10 12

alcohol chain length

FIG. 5 Limiting alcohol concentrations for foam to exhibit instability at various


concentrations of a blend of ethoxylated nonyl phenols (in 0.1 M KC1): (1) 1%, (2)
0.1 %, (3) 0.5%, (4) 0.025%, (5) 0.005%, (6) 0.001 %. (From Ref. 34.)

oil lens from the air-water surface in the foam film (see Fig. 2) breaks.
Unfortunately, Kruglyakov [34,35] does not state precisely how the bridging
oil lens formed by rupture of that asymmetric film can give rise to foam
film collapse. It is, however, possible to deduce a detailed hypothesis as we
describe below (see Sec. V.C.).
In interpreting the results shown in Fig. 5, Kruglyakov and Koretskaya
[34,35) note that for all the alcohol-surfactant solution combinations the final
entry coefficient E" is everywhere positive. Ineffective antifoam behavior
18 Garrett

• •
(a)

• (b)

FIG. 6 Formation of "asymmetrical film" where oil lens is present in foam film.
(From Refs. 34 and 35.)

for the lower-chain-length alcohols is ascribed to the higher solubility of


these materials. Declining effectiveness of the higher-chain-length alcohols
is ascribed to decline in polarity and increase in surfactant adsorption con-
spiring to enhance the stability of the oil-water-air films. That the optimum
alcohol chain length for antifoam effectiveness declines with increasing sur-
factant concentration is then attributed to a concomitant increase in surfac-
tant adsorption at the alcohol-water surface. This is supposed to enhance the
stability of the oil-water-air film which diminishes the effectiveness of the
longer-chain-length alcohols. Unfortunately Kruglyakov and Koretskaya
[34,35] give no direct experimental evidence to confirm that the stability of
the asymmetric films changes with alcohol chain length in a manner which
correlates with changing antifoam effectiveness. Clearly, study of the ki-
netics of the process of emergence of alcohol droplets into the relevant air-
water surfaces would be instructive.
It is argued by Kruglyakov [34] and others [36,37] that the well-known
phenomenon of reduction in foamability of ethoxylated nonionic surfactants
at temperatures above the cloud point (see [38), for example) may also con-
cern the formation of oil lenses by nonionic-rich cloud phase droplets in a
manner analogous to that proposed for the alcohols. Clear evidence that such
Mode of Action of Antifoams 19

droplets contribute to an antifoam effect is presented by Koretskaya [37],


who showed that removal of the cloud phase droplets by filtration restores
the foamability despite the reduction in overall surfactant concentration caused
by such a procedure. The effectiveness of surfactant antifoams based upon
such ethoxylated (or propoxylated) nonionics may therefore owe much to a
low tendency of any other surfactant present to adsorb on the surface of
cloud phase droplets. However, direct experimental evidence is lacking.
Jn summary, then, it would seem that spreading to form duplex films is
not a necessary property of antifoam oils. Indeed, Kruglyakov and Koret-
skaya [34, 35] advance an alternative hypothesis which requires only that
the entry coefficient be positive so that the oil may form lenses. This hy-
pothesis is more general in that such lenses will often form even if the oil
exhibits a positive initial spreading coefficient.
An interesting aspect of the hypothesis concerns the supposed effect of
adsorbed surfactant on the stability of asymmetrical oil-foaming liquid-air
films when that liquid is water. A consequence of this hypothesis is that if
the oil is nonpolar with a high interfacial tension against water, then sur-
factant adsorption will be high and antifoam efficiency low even if E > 0
or E" > 0. Hydrocarbons are examples of such nonpolar oils which are
known to be relatively ineffective when used alone in aqueous ionic sur-
factant solutions [39 ,40]. The anti foam effectiveness of these oils in such
solutions can, however, be enhanced significantly by addition of hydropho-
bic particles to the oil [39 ,40]. It would not be inconsistent with the hy-
pothesis of Kruglyakov and Koretskaya [34,35) if we were to suppose that
the role of such particles concerns the stability of the relevant oil-water-air
films. We address the possibility in more detail below (see Sec. V.E.).

C. Surface Tension Gradients and Theories of


Antifoam Mechanism in Heterogeneous Systems
We have disposed of the proposition that spreading from oil droplets to form
duplex films is a necessary aspect of anti foam mechanism. However, certain
undissolved materials which do spread to form duplex films (or even mono-
layers) may function because of that property. Here we examine two dif-
ferent approaches to antifoam mechanism which require such behavior. The
first concerns elimination of stabilizing surface tension gradients as a result
of displacement of the stabilizing surfactant monolayer by a spread film of
antifoam. The second concerns the effect of the surface tension gradient,
caused by spreading, on the thinning rate of foam films.
1. Elimination of Surface Tension Gradients
Kitchener [27] has argued that the role of "a completely effective antifoam
agent must be to eliminate surface elasticity i.e. to produce a surface which
20 Garrett

has substantially constant tension when subjected to expansion." In effect,


this is a corollary of the argument of Lucassen [2] (which we have outlined
in Sec. II.A.) concerning the role of surface tension gradients in stabilizing
foam films against external stress. It is also similar to the argument stated
by Burcik [25]. As we have seen, Ross and Haak [21) have applied a vari-
ation of it to interpretation of antifoam behavior in homogeneous systems.
Here we examine the application of this argument to heterogeneous systems,
i.e., systems where undissolved antifoam materials are present.
Elimination of elasticity in heterogeneous systems is most probable when
oil lenses in the surface are in equilibrium with a monolayer contaminated
with the oil. Expansion of the surface will then mean more oil spreading
out from the lenses, which will tend to maintain the surface tension. A
monolayer contaminated with oil in equilibrium with oil lenses is of course
the probable consequence of disproportionation of a nonequilibrium duplex
film [32,33]. Obviously the effectiveness of the overall process will be de-
termined by rate effects. Interpretation of measurements of the effect of an-
tifoam materials on the relaxation of surface tension gradients therefore re-
quires knowledge of the time scale of the relevant effects in foams.
Abe and Matsumura [41) have assessed the supposed role of the elimi-
nation of surface tension gradients under dynamic conditions in determining
heterogeneous antifoam effectiveness. This work concerned the antifoam ef-
fect of alcohols on the foam behavior of an aqueous solution of sodium
dodecylbenzenesulfonate. The alcohols included normal alcohols, branched
alcohols, and diols.
These authors followed Ross and Haak [21] and measured the dynamic
surface tension of dispersions of alcohols in solutions of sodium dodecyl-
benzenesulfonate, using the oscillating jet method. Here we remember that
if the surface tension of the foaming liquid is lowered rapidly by the anti-
foam, then any rapid stretching of a foam film element will not result in an
increase in surface tension of that element. There will therefore be no surface
tension gradient between that element and the adjacent thicker elements. No
reinforcing flow of liquid into the thin element will then occur. It is also
obvious that the film will not be able to effectively withstand the effects of
any external stress (due to gravity for example).
The limitations of this approach with respect to the time scale of the
measurement of dynamic surface tension have already been outlined above
(Sec. III.A.). Despite these limitations Abe and Matsumura [41] show a
correlation between the dynamic surface tensions at surface ages of 2 x 10- 2
s and foamability in the case of the normal alcohols. Thus, under these
conditions the optimum antifoam is octanol, which produces the most rapid
rate of surface tension reduction, as indicated by dynamic surface tension
measurements. We should note here though that this correlation concerns
Mode of Action of Antifoams 21

measurement of dynamic surface tensions in the presence of 7 .6 x 10- 3 M


alcohol concentration for comparison with foamability in the presence of
3.84 x 10- 2 M alcohol (i.e., 5 g dm- 3 in the case of octanol). However,
at 7. 68 x 10- 3 M the octanol is essentially ineffective as an antifoam (see
Fig. 23 in Ref. 41) presumably because it is below the solubility limit.
Moreover, increase in the octanol concentration above 7 .68 x 10- 3 M has
little effect on the dynamic surface tension at a surface age of 2 x 10- 2 s.
All of this would tend to suggest that whatever the significance of the cor-
relation it has limited relevance for the basic proposition that rapid reduction
of dynamic surface tension over the time scales measured by the oscillating
jet will reduce foamability. Abe and Matsumura [41] in fact demonstrate
the absence of any correlation between the rate of surface tension lowering
(albeit again with different concentrations between surface tension and foam-
ability measurements) for alkanediols. Here the optimum antifoam effec-
tiveness is found with those compounds which yield the slowest rate of sur-
face tension reduction.
Curiously, both Abe and Matsumura [41] and Kruglyakov and Koret-
skaya [35] have studied the effects of normal alcohols on the foam of aqueous
sodium dodecylbenzenesulfonate solutions. Both find octanol to have the
optimum chain length for antifoam effect at surfactant concentrations in the
region of 1 g dm- 3 (albeit at different temperatures and ionic strengths).
They also find that the antifoam concentration for significant effect under
these conditions is extremely high relative to that of the surfactant-in the
region of 5-10 g dm - 3 for octanol. Despite consensus concerning these ex-
perimental findings, the two groups of authors have adopted completely dif-
ferent hypotheses describing the mode of antifoam action of these alcohols.
Surprisingly, Abe and Matsumura [41] do not cite the earlier work of Krug-
lyakov and Koretskaya [35].
The argument concerning elimination of surface elasticity (and therefore
smface tension gradients) by antifoams has been extended to nonaqueous
foaming liquids by Callaghan et al. (42]. These workers have studied the
effect of polydimethylsiloxanes on the surface elasticity of crude oil. Poly-
dimethylsiloxanes are used as antifoams to assist gas-oil separation during
crude oil production and are apparently effective at the remarkably low con-
centration of 1 part in 106 (which presumably still exceeds the solubility
limit). Callaghan et al. [42] find that polydimethylsiloxane diminishes the
frequency-dependent dynamic dilational (elastic) modulus lei = d'Y Ao(t) / d
In A(t) relative to that found for the uncontaminated oil. Here "Y Ao(t) is the
time-dependent air-crude oil surface tension, and A(t) is the area of a con-
strained element of air-crude oil surface subject to time-dependent dilation.
The effect is more marked the higher the molecular weight (or viscosity) of
the polydimethylsiloxane. This correlates with an enhanced antifoam effec-
22 Garrett

tiveness found with increase in molecular weight. Callaghan et al. [42] argue
that all of this represents evidence that the polydimethylsiloxane functions
by elimination of surface elasticity and tension gradients. However,
McKendrick et al. [43 j find a frequency-independent increase in dynamic
dilational modulus of a hydrocarbon oil upon addition of a fluorosilicone
antifoam. Moreover, the correlations of Callaghan et al. [42] concern ex-
tremely long time scales in the range 10-103 s. Whether these correlations
are fortuitous or represent clear evidence for the proposition under test re-
quires additional information concerning the relevance of those time scales.
Absence of hard evidence concerning the magnitude of the relevant time
scales is thrown into high relief if comparison is made between the ap-
proaches of Abe and Matsumura [411 and Callaghan et al. [42]. These work-
ers have looked at time scales which are separated by more than three orders
of magnitude. This is surprising, since they are both seeking evidence for
the same basic proposition concerning the elimination of surface elasticity
in foam films by antifoam oils.
2. Surface Tension Gradients Induced by Spreading
Antifoam
We have seen that some authors form the view that antifoams eliminate
stabilizing surface tension gradients in foam films. Here we consider the
proposition that antifoams cause destabilizing surface tension gradients in
foam films.
The proposed mechanism concerns spreading from an antifoam source
present on the surface of a foam film. Such spreading is driven by a surface
tension gradient between the spreading source and the leading edge of the
spreading front. This surface tension gradient will act as a shear force drag-
ging the underlying liquid away from the source to supposedly cause cata-
strophic thinning and foam film rupture. The process is shown schematically
in Fig. 7. It is clearly a consequence of the well-known Marangoni effect,
and we refer to it below as Marangoni spreading.
This mechanism was first proposed by Ewers and Sutherland [26]. These
authors identified three categories of antifoam substance which could, in
principle, function by this mechanism. These are:
"solids or liquids containing surface active material other than the substance
stabilizing the film"
"liquids which contain the foam stabilizer in higher concentration than it is
present in the foam"
"vapours of surface active liquids" (including the well-known example of
the adverse effect of ether vapor on foam stability)
In the first category Ewers and Sutherland [26] included all spreading oils
for which S > 0. In addition, they included monolayer spreading where a
Mode of Action of Antifoams 23

Spreading from
antifoam particle

--
drags
underlying liquid
"\ air
~---

water

air

l film thins in vicinity


of particle

FIG. 7 Foam film rupture caused by spreading from antifoam particle.

solid or liquid forms a monolayer for which the initial spreading pressure
Sm is
(13)

where "Im is the equilibrium surface tension of the spreading monolayer on


a surface of the foaming liquid from which surfactant has been removed.
The underlying driving force for all of these categories of antifoam sub-
stance is the surface tension gradient dS / dx or dSm/dx, where x is the spreading
distance. Therefore, according to Ewers and Sutherland [26] the greater the
24 Garrett

magnitude of S (or Sm) the greater the force and the more effective the an-
tifoam. Clearly here S could be measured on a nonequilibrium foam film
Where ("Y AF)dynamic > ("Y AF)equilibrium ·
This approach of Ewers and Sutherland [26] was advanced with some
claim to wide generality. However, the finding that many substances yield
antifoam effects without causing destabilizing surface tension gradients
[19,22,29,31,34,44] confounds exclusive generality. Moreover, a difficulty
with the mechanism concerns the implicit assumption that the antifoam is
always to be found in the thinnest and most vulnerable part of the foam.
Thus, if the antifoam entity spreads from the plateau border, this will drag
liquid into the adjacent foam films, which will stabilize those films by in-
creasing their thicknesses.
Another limitation of the approach of Ewers and Sutherland [26] is that
it is essentially qualitative. However, there have been some attempts to put
it on a quantitative basis [ 17 ,45]. The first such attempt was due to Shearer
and Akers [ 17], who were concerned with the antifoam effect of polydi-
methylsiloxanes on lube oils. Here the effect of antifoam on a monolayer
raft of bubbles is modeled theoretically and the results compared with
experiment.
In this approach, spreading from droplets of antifoam on the foaming
liquid surface is supposed to augment gravity drainage from foam films.
This gravity drainage is modeled by assuming that the boundaries of the
foam films are immobile, in that Poiseuille flow prevails (with a parabolic
velocity profile). A schematic illustration of the model where bubbles are
supposed to form a raft at the surface of the foaming liquid is shown in Fig.
8a. Shearer and Akers [ 17] calculate that the time t 8 taken for a thick foam
film surrounding a bubble to drain to the thickness hcrit• at which rupture
occurs, is
18T)FRb
fg ~ 2 (14)
pghcrit

Here p and TJF are respectively the density and viscosity of the foaming
liquid, and Rb is the radius of the bubble. Equation 14 implies that the time
t 8 is independent of the height of the film above the planar liquid surface at
which the bubble is resting. In actuality the film would become relatively
thin at the top of the bubble. Failure to account for this effect is a conse-
quence of incomplete consideration of continuity by Shearer and Akers [ 17]
in the derivation of Eq. 14.
The treatment of film draining due to spreading given by Shearer and
Akers [ 17] is summarized here to illustrate the essentials of the argument.
Thus, consider one polydimethylsiloxone oil droplet of diameter a at the
center of the foam film as indicated in Fig. 8b. The total spreading force is
Mode of Action of Antifoams 25

parabolic velocity
profile in film

(a)

velocity of antifoam droplet


/
spreading monolayer
=U

spreading monolayer

(b)

FIG. 8 Model of Shearer and Akers [17] describing effect of gravity drainage and
spreading from an antifoam droplet on film thinning. (a) Foam film drainage under
gravity from a bubble at a planar liquid surface (where h ~ Rb). (b) Foam film
drainage induced by spreading from antifoam droplet.

assumed to be '!TaS at the perimeter of the oil droplet, where a is the droplet
diameter. It is also assumed that S becomes zero at the edge of the bubble
in the plane of the liquid surface so that a linear gradient in S exists with
respect to radial distance from the antifoam droplet. The shear stress ,. is
therefore crudely approximated as
26 Garrett

aS
-r=-- (15)
k.R~
where k. is a geometrical constant which describes the area of the bubble
surface above the plane of the liquid surface. This model of the shear stress
obviously ignores one of the essential features of the spreading process,
namely that the spreading front will expand over the substrate and the shear
stress will decline with time. It will therefore tend to exaggerate the effects
of spreading.
Shearer and Akers [ 17 J argue that a linear velocity gradient u / h will exist
in the film due to the imposed shear stress. Here h is the film thickness,
and u is the velocity of the spreading surface (which is assumed to be con-
stant throughout drainage). The opposite surface of the film is assumed to
be stationary where u=O (see Fig. Sb). Therefore it is possible to deduce
from Eq. 15 that
aSh
u=--- (16)
ka'YJFR~
Shearer and Akers [ 17] use a crude averaging procedure to allow for con-
tinuity in the derivation of Eq. 14. They also use a similar procedure to
deduce from Eq. 16 that the time t. taken for spreading to thin a thick film
down to the critical rupture thickness he,,, is given by
kb'YJFR~
ta= - - (17)
aShcrit
where kb is a constant. Using parameters relevant for typical polydimethyl-
siloxane/lube oil combinations, Shearer and Akers [ 17] show that ta ~ Iv
so that any contribution of gravity drainage to ta may be neglected. Indeed
their calculations suggest that t./tg could be of order 10- 3 . If the crude as-
sumptions upon which they are based are not too misleading, then this would
imply that spreading from polydimethylsiloxane droplets could significantly
increase the rate of foam film rupture in lube oils.
Shearer and Akers [ 17] allow for the possible presence of several droplets
of antifoam in a foam film by simply increasing the supposed spreading
perimeter to mra, where n is the number of droplets per film. If, however,
the droplets are evenly dispersed over the surface of foam films, then spread-
ing will occur in opposing directions. Flow of liquid from the film will be
less than that predicted if the spreading perimeter is simply increased to mra.
Equations 14 and 17 may be combined to give
t.S kcR~hcrit
(18)
Mode of Action of Antifoams 27

where kc is a constant and where allowance is made for the presence of n


droplets in the film. The group t.S /tgp is seen to be independent of the prop-
erties of the foaming liquid provided the bubble size can be controlled and
hcrit is invariant. Indeed, log/log plots of this group against antifoam con-
centration for three different lube oils with a given polydimethylsiloxane
were shown by Shearer and Akers fl 7) to lie on the same curve. Here tg
and ta were obtained experimentally by assessing the half-life of bubble rafts.
All of this is offered as experimental evidence for the validity of Eq. 18 and
the model from which it is derived. In view of all the assumptions implicit
in the approach of Shearer and Akers [ 17], this experimental accord with
theory is remarkable.
Prins [45] has adopted a different approach to the theoretical description
of the effect of spreading from heterogeneities on foam film stability. In
contrast to Shearer and Akers [ 17] he recognizes that the spreading process
involves a spreading front which advances over the substrate surface. This
process has been the subject of a number of independent studies. Thus, Fay
[46] and Hoult [47] show from dimensional analysis that an oil (for which
S > 0) will spread as a duplex film over a clean water surface in a linear
manner at a rate given by
2 ) 1/4
x = k, ( __!___ t3/4 (19)
'lJwPw
where x is the spreading distance, k, is a constant which has a value of about
1.33 for liquids such as polydimethylsiloxanes of viscosity :s 1000 mPa s
[48], and 'llw and Pw are, respectively, the viscosity and density of water.
The rate of spreading is seen to be dependent upon S but is independent of
the viscosity of the oil. Joos and Pintens [49] treat the process as a longi-
tudinal disturbance and deduce theoretically that k, = 1.15.
As we have seen, the spreading antifoam oil will cause the underlying
liquid to move in the same direction. lf the depth to which this movement
penetrates is 8, we can write approximately
s dx/dt
- = 'llw_o_ (20)
x
Substituting Eq. 19 in Eq. 20, using the value k, = 1.15 given by Joos and
Pintens [49], yields for the penetration depth
1/2
'YJwf )
8= -
(
(21)
Pw
Prins [45] supposes that foam film collapse will occur when the penetration
depth of a spreading duplex film equals the thickness of the foam film over
28 Garrett

which it is spreading. Clearly the depth will increase with time (eq. 21),
and so some estimate of maximum time available is necessary in order to
assess the maximum penetration depth &m. Prins estimates the maximum
extent Xm to which the spreading may occur in a radial sense from
-rra3
- = -rrx2m dm (22)
6

where dm is the thickness of the spreading film at the maximum extent and
a is the diameter of the oil droplet. The time of spreading is then obtained
somewhat approximately by substituting Xm, derived for radial spreading,
into Eq. 19, which is derived for linear spreading. Thus, Prins [45] combines
Eqs. 19, 21, and 22 to obtain the maximum penetration depth:

(~2 )1/3
8m=0.5a (23)
Spwdm
If foam film collapse occurs when the penetration depth equals the thickness
of the film, then the larger the value of &m the thicker the films which may
be ruptured. Prins [45] therefore equates high values of &m with high prob-
ability of foam film rupture. Equation 23 therefore implies that antifoam
effectiveness will increase with increasing size of the antifoam droplet.
At least in this respect there is agreement with Shearer and Akers [ 17], al-
though for different reasons (see Eq. 18). Prins [45) has shown that increas-
ing droplet size can increase antifoam effectiveness for small droplet sizes.
However, this does not continue indefinitely because as the droplet size be-
comes large the probability of an antifoam entity being present in a foam
film declines. An optimum particle size for high antifoam effectiveness
is therefore obtained. Experimental results for soya bean oil antifoam in
sodium caseinate solution illustrate this behavior and are reproduced in
Fig. 9.
The crude nature of the model presented by Prins [45] gives rise to an
awkward feature in Eq. 23. Thus, if S = 0 then we appear to have &m =
oo. Equation 23 in fact means that antifoam effectiveness is predicted to
increase as S decreases. Thus, Prins [45] presents a model of antifoam ac-
tion, based upon spreading driven by surface tension forces, which predicts
increasing antifoam effectiveness as those forces become weaker. Intuition
would suggest an opposite conclusion (which is stated by Ewers and Suth-
erland [26] and deduced by Shearer and Akers [ 17)).
We are therefore left without an entirely satisfactory mathematical treat-
ment of the antifoam mechanism proposed by Ewers and Sutherland [26].
Clearly the problem is difficult. However, a solution would permit more
serious assessment of the significance of the mechanism particularly with
Mode of Action of Antifoams 29

Foam
half-life time • 0.1% oil by
minutes volume
8
0 0.15% oil by
volume

2 4 6
a I microns

FIG. 9 Foam half-life time of 0.05 wt.% sodium caseinate solutions as a function
of mean drop diameter a of soya bean oil. (From Ref. 45, reproduced by permission
of the Royal Society of Chemistry).

regard to the effect of the magnitude of S (or Sm). Such a solution should
take account of the effect of compression of the surfactant monolayer at the
foaming liquid-air surface by the spreading front. This effect will clearly
reduce the rate of spreading and will diminish the threat to foam film stability.
There are other problems with the mechanism. Spreading from antifoam
entities present in the plateau borders could in principle reinforce foam film
stability. Spreading from many different sources in the surface of a foam
film would presumably rapidly eliminate surface tension gradients so that
rupture induced by surface-tension-driven flow would become unlikely.
30 Garrett

However, elimination of surface tension gradients can also in principle con-


tribute to diminished foam film stability.
Another problem concerns the nonequilibrium nature of the process of
spreading. Thus, if the antifoam and foaming liquid are mutually saturated,
we can have S" < 0 (and Sm < 0) [32,33]. Spreading will then only occur
under nonequilibrium conditions where, for example, the surface tensions
of foam films are markedly higher than equilibrium as a result of slow sur-
factant transport to the relevant surfaces.
We therefore have a mechanism which relies on a necessarily transient
process. However, materials which have the relevant properties, such as
spreading oils, will also be capable of participating in other mechanisms.
The latter include elimination of surface tension gradients and formation of
unstable bridging oil lenses (see Sec. V. C.). Unequivocal experimental ver-
ification of a role for this mechanism will therefore represent a formidable
challenge.

IV. INERT HYDROPHOBIC PARTICLES AND


CAPILLARY THEORIES OF ANTIFOAM
MECHANISM FOR AQUEOUS SYSTEMS
A. Early Work
Solid particles which adhere to fluid-fluid surfaces so that part of their sur-
face is exposed to one fluid and part to the other fluid exhibit a finite contact
angle at the relevant surface. For almost three-quarters of a century it has
been known that such particles may have a drastic effect on the behavior of
dispersions formed by mixing fluids. One of the earliest publications is that
of Pickering concerning the effect of particles on oil-water emulsion behav-
ior [50]. A few years later Bartsch [51] described the stabilizing effect of
hydrophobed mineral particles on froths fo1med by aqueous solutions of
3-methylbutanol.
Since then there have been many publications concerning the effect of
particles on the formation and stability of foams, froths, and emulsions.
Here we are mainly concerned with the destabilizing effect of particles on
foams and froths. Practical examples include the effect of hydrophobed min-
eral particles and collector precipitates on the stability of mineral flotation
froths [52,53]. Another is the use of wax (541 to control the foam of de-
tergent formulations for automatic washing machines.
Arguably the earliest suggestion that an adverse effect of particles on
aqueous froths may be attributable to low wettability (or high contact angle
at the air-water surface, measured through water) was made by Mokrushin
40 years ago [55]. This concerned the effect of metal sulfides on froths
Mode of Action of Antifoams 31

formed by gelatin solutions. A few years later Dombrowski and Fraser [56]
reported direct observation of the effect of wettability of particles on the
disintegration of thin films of water or alcohol formed from spray nozzles.
Here particles which were not wetted by the film liquid caused perforation
of the film when the particle size was of the same order as the film thickness.
Particles which were wetted by the film liquid had no effect on the manner
of disintegration.
A later reference to the role of the wettability (or the contact angle) of
dispersed solid material in determining froth stability is due to Livshitz and
Dudenkov [57]. Here the destabilizing effect of hydrocarbon particles on
froths is attributed to air-water contact angles >90° by analogy with the
effect of particles on oil-water emulsion behavior. Thus the presence of par-
ticles with a contact angle >90° at the oil-water surface (measured through
the water) in an oil-water mixture will produce a water-in-oil emulsion [58].
Livshitz and Dudenkov [57] argue that analogous behavior in the case of an
air-water system is tantamount to destruction of the foam. This of course
gives no mechanism for either supposedly analogous phenomenon. How-
ever, in a later paper, concerned with the effect of hydrophobed metal xan-
thates and oleate precipitates on flotation froths, Livshitz and Dudenkov [59]
suggest that hydrophobic particles may bridge aqueous films. The more hy-
drophobic the particle, the closer are the wetting perimeters of each air-water
surface at the particle. This will supposedly reduce the thickness of the film,
which will result in acceleration of film rupture. In the case of a hydrophilic
particle, on the other hand, contact angle formation does not take place and
there is no film rupture. This mechanism appears to suggest that there should
be a continuous improvement in defoaming ability with increasing contact
angle.
A difficulty with establishing the role of wettability in determining the
effectiveness of particles in breaking foams and froths is measurement of
the relevant contact angle. This is particularly so for precipitated particles
of sparingly soluble materials. Thus, representative smooth surfaces upon
which contact angles can be readily measured may not be easily prepared.
In recognition of this Dudenkov suggested that the solubility products of
polyvalent metal salt precipitates of butyl xanthogenates and oleates should
correlate with the contact angles (60]. Thus, the lower the solubility product
the more hydrophobic is the precipitate for a given anionic hydrophobe.
Whatever the merit of this supposed correlation, Dudenkov [60] produces
results which indicate a sharp increase in the volume of air present in a froth
for precipitates of oleate or xanthogenate with solubility products greater
than a critical value. The latter decreases with increase in frother concen-
tration. Examples of their results are presented in Fig. 10. These are inter-
preted by Dudenkov [60] as indicating the importance of wettability in de-
32 Garrett

Air
Volume
in Froth
cm3
450

ZnX
400 (a) Metal xanthogenates

350

300

250

-35 -30 -25 -20 -15 -10 -5


Air
Volume IOQ10 (solubility product)
in Froth
cm3 150
(b) Metal oleates

100


PbOL2
50
FeOL3
&-~--.:.~~~~....... caOL2

I
-30 -25 -20 -15 -10 -5
IOQ10 (solubility product)

FIG. 10 Relationship between air volume in froth and solubility products of hy-
drophobic precipitates. (From Ref. 60.) (a) Metal butyl xanthogenates at 5 x 10- 5
Min 30 mg dm- 3 commercial frother solution. (b) Metal oleates at 2 x 10- 5 Min
10 mg dm- 3 commercial frother solution. (Reproduced by permission of Primary
Sources.)
Mode of Action of Antifoams 33

termining antifoam effectiveness. But the solubility products given by


Dudenkov [60] may be suspect. Thus, the value given for calcium oleate is
almost two orders of magnitude lower than that given by Irani and Callis
[61].
Similar observations have been reported by Peper [62], who considered
the effect of sodium soaps and fatty acids on the foamability and foam sta-
bility of dilute solutions of anionic surfactant solutions. Peper [62] found
that in the presence of a stoichiometric excess of calcium chloride (for cal-
cium soap formation) the adverse effect of sodium soaps on foam was in
the order sodium stearate = sodium palmitate > sodium oleate > sodium
laurate, which follows the order of solubility products K.P of the correspond-
ing calcium soaps (calcium stearate, K.P - 2 x 10- 20 ; calcium palmitate,
K,P - 6 X 10- 18 ; calcium oleate, Ksp - 10- 15 ; calcium laurate, K.P - 6 X
10- 13 at 25°C [61]). Curiously Peper [62] interprets his results in terms of
formation of rigid islands of calcium soap monolayer interspersed with gas-
eous film of adsorbed surfactant. Peper [62] asserts that these islands will
make the film unstable because of their "inflexible brittle nature." No the-
oretical arguments are given for why these should be unstable. Any role for
the calcium soap precipitates, which will undoubtedly be found under these
conditions, is ignored.
Peper (62] found that fatty acids reduce the surface tension of solutions
of sodium dodecylbenzenesulfonate containing calcium chloride. The effect
was found to be least pronounced for the fatty acid which forms the least
soluble calcium soaps and the best antifoam. It is difficult to reconcile this
finding with a spreading mechanism for antifoam action (see Sec. III.C.2.).
B. Contact Angles and Particle Bridging Mechanism
Attempts to relate the foam behavior of hydrophobic particles directly to
contact angles followed more than a decade after the work of Dudenkov
[60]. Of these a paper by Garrett [44] concerned finely divided polytetra-
fluoroethylene (PTFE) as antifoam material. PTFE has the merit of inertness
so that any possible contribution from Marangoni spreading to the antifoam
effect can be unambiguously eliminated. Moreover, reliable representative
contact angles can be measured on polished PTFE plates.
The effect of finely divided (-5 µm) PTFE particles on the volume of
foam generated by cylinder shaking for several aqueous surfactant solutions
is reproduced in Fig. 11. An apparent correlation between volume of foam
destroyed and receding contact angles is revealed. That the effects do not
concern adsorption loss of surfactant onto PTFE surfaces can easily be dem-
onstrated by filtering off the PTFE. Foamability of the solution is restored.
In seeking an explanation for the antifoam behavior of PTFE which is
related to wettability, Garrett [44] followed Livshitz and Dudenkov [59] and
34 Garrett

Volume Foam
Destroyed
cm 3
120

100

80

60

40

20

10 20 30 40 50 60 70 80
Contact Angle OAw I degrees

e 5gl - 1 detergent product O 0.01M sodium dodecyl sulphate I


0.1MNaCI

• 0.01 M sodium dodecyl sulphate o 0.002M tetradecyldimethyl-


aminopropane sulphonate

FIG. 11 Antifoam effect of polytetrafluorcthylene particles as a function of reced-


ing contact angle. (From Ref. 44, reproduced by permission of Academic Press Inc.)

considered a bridging hydrophobic particle. Depending on the contact angle,


two consequences may be distinguished for a hydrophobic spherical particle
emerging into both air-water surfaces of a foam film. Thus, if the contact
angle is >90°, there is no condition of mechanical equilibrium available to
the particle. There will be an unbalanced capillary pressure tlP in the vi-
cinity of the particle given by the Laplace equation

l:::..P = 'YAw(2_ +
R1
2-)
Rz
(24)

where R 1 and R 2 are the two radii of curvature describing the curved air-
water surface in the vicinity of the particle and 'Y Aw i.s the air-water surface
tension. The configuration is shown in Fig. 12 where the particle is assumed
to be small so that the effect of gravity upon the shape of the air-water
surface may be neglected. The two radii of curvative are of opposite sign
so that R 1 < 0, and therefore the direction of the unbalanced capillary pres-
sure depends upon the relative magnitudes of R 1 and R 2 • The ratio R 2 /R 1 is
not obvious by inspection, although it has been simply assumed in later work
Mode of Action of Antifoams 35

/ '

___.__./
\
8Aw I >
' ;.........._
... \ ... ;..
7
------,.~ '< -~----

' foam film drainage


rate increased by
' , capillary pressure

·+
particle completely
dewetted to form hole
in foam film

FIG. 12 Foam film rupture caused by a spherical particle with contact angle 0Aw
> 90°.

by Frye and Berg [63] that

Rz
-<:iS I (25)
R,
so that Eq. 24 becomes

(26)

The capillary force will therefore act in a direction to cause enhanced drain-
age out of the foam film. The process of drainage will continue until the
two three-phase contact lines at the particle surface are coincident, where-
36 Garrett

upon a hole will form in the foam film. It has been shown by de Vries [64]
that if the hole is sufficiently large it will spontaneously increase in size and
lead to film collapse.
The second consequence of a particle emerging into both air-water sur-
faces of a foam film follows if the contact angle <90°. The resulting con-
figurations are depicted in Fig. 13. Here the initial configuration is similar
to that found for particles with contact angles >90° where, if inequality 25
is valid, an unbalanced capillary pressure will enhance drainage out of the
foam film. However, as the film drains it will always attain some thickness
where the film is planar and where the capillary pressure becomes zero.
Further drainage will then result in a change in sign of both radii of cur-
vature. If condition 25 is assumed, this will mean an unbalanced capillary
pressure inhibiting further drainage of the film. It is therefore possible that
a particle with this configuration will actually stabilize foam films. That
hydrophobic particles may stabilize foams has in fact been occasionally re-
ported in the literature [52,57 ,65,66].
The essentials of the mechanism of film collapse by spherical particles
have been confirmed in a cinematographic study by Dippenaar [52], using
hydrophobed glass spheres. Representations of high-speed movie frames are
reproduced in Fig. 14 for a glass bead of 250 µm diameter and a contact
angle of 102° in a distilled water film. Here the contact angle formed initially
at the lower air-water surface is seen to be apparently less than the equilib-
rium value. The usual interpretation of such dynamic contact angles is that
the equilibrium contact angle always prevails near the three-phase contact
line and that the air-water surface exhibits extreme curvature in the same
region to give an apparent macroscopic contact angle much lower than the
equilibrium value. It is the capillary pressure implied by that curvature which
produces the force driving film thinning. Collapse occurs about an order of
magnitude faster than would have occurred if the particle had been absent
[52]. This is seen to occur as the two three-phase contact lines become coin-
cident. By contrast, a sphere of contact angle 74° had no effect on film
rupture times and adopted a configuration similar to that of Fig. 13c.
A basic difficulty with application of this mechanism of collapse by hy-
drophobic particles to the results shown in Fig. 11 is that the contact angle
for the PTFE particles is always <90°. However, the contact angles cited
in Fig. 11 are equilibrium values. Foam generation is a nonequilibrium phe-
nomenon where air-water surfaces will suffer rapid overall expansion. This
will mean that the surface tension of a freshly generated foam film will in
general be greater than the equilibrium surface tension of the solution from
which it is formed because adsorption of surfactant is not instantaneous. This
in tum should mean that the actual contact angle prevailing during foam
generation will be greater than the equilibrium contact angle. Since the re-
Mode of Action of Antifoams 37

\ 8Aw)
'\ .;

< \

+
(a)

(b) \
air-water surface everywhere
planar so capillary pressure due
to particle is zero

'\ \

.....
~--
(c)
foam film drainage
rate decreased by
capillary pressure

I
I
CJ

FIG. 13 Effect of spherical particle with contact angle 0Aw < 90° on foam film.
38 Garrett

Time; ms
70

13,5

14,7

24,5

116

FIG. 14 Reproduction of high-speed cinematographic film frames showing inter-


action of hydrophobic glass bead (0Aw = 102°) with an aqueous film. (From Ref.
52, reproduced by permission of Elsevier Science Publishers).

ceding contact angle for PTFE against distilled water is in the region of 95°
[44J, dynamic contact angles >90° are in principle possible. However, such
values would only be attained if all surfaces were almost denuded of sur-
factant. Survival of films with such surfaces for sufficient time to form a
foam would seem unlikely anyway regardless of the presence of particles.
These considerations suggest that contact angles >90° are unlikely to have
prevailed in the foam experiments from which the results shown in Fig. 11
are taken.
In an attempt to explain the discrepancy, Garrett [44] suggested that the
contact angle required for film collapse by a particle may be determined by
Mode of Action of Antifoams 39

\
\

Helmholtz Free Energy


Change< 0

14 - - - - -
... am - - - - - - - l i t
FIG. 15 Effect of particle geometry on foam collapse; disk-shaped geometry with
various aspect ratios a. (From Ref. 44, reproduced by permission of Academic Press
Inc.)

particle geometry. Thus, PTFE particles are modeled as disk shapes (with
round edges) of various aspect ratios. Hole formation in a foam film by
dewetting of particles is assumed to occur by the model shown in Fig. 15.
The condition for hole formation is simply that the Helmholtz free-energy
change accompanying the dewetting process shown in Fig. 15 should be
negative. It follows from this model that the contact angle for film collapse
decreases with increasing aspect ratios of the particles L44]. Thus, if the
aspect ratios are sufficiently large, particles with contact angles <90° are
predicted to cause film rupture by this mechanism. A major flaw with this
approach is, however, neglect of the effects of viscous dissipation. Thus,
for contact angles <90° an apparently stable configuration analogous to that
shown in Fig. 13b is available for the particle. That hole formation is pre-
dicted despite this is due to the assumption that the free-energy gain in ar-
riving at that configuration is all then available for doing work to produce
complete dewetting of the particle.
A superior explanation for the role of particle geometry in producing foam
collapse with contact angles <90°C emerges from the cinematographic study
40 Garrett

air
water

(a) Orientation if 0 < 8Aw < 90°

(b) Orientation if 45° < 8Aw < 135.0

FIG. 16 Orientations of cubic particle at air-water surface.

by Dippenaar [52]. Here Dippenaar [52] examined the behavior of cubic


particles of xanthanated galena with a contact angle of 80 ± 8° against water.
Such particles were observed to adopt two different orientations, with about
equal probability, at the air-water surface. These orientations are depicted
in Fig. 16, where it is seen that the air-water surface intersects the particle
edges for both orientations.
The behavior of air-water surfaces at edges is usually interpreted by sup-
posing that the equilibrium contact angle is always satisfied at the submi-
croscopic level [67]. The edge is represented as having a submicroscopic
circular cross section [67). The air-water surface may then adopt an apparent
single macroscopic angle at the edge for a range of contact angles, any of
which may be satisfied at the submicroscopic level. All of this is illustrated
in Fig. 16, where it is clear that the two orientations for a cubic particle
require contact angles 0 < 6Aw < 90° and 45 < 6Aw < 135°, so that the
Mode of Action of Antifoams 41

xanthanated galena used by Dippenaar [52] with a contact angle of 80° may
adopt either.
Dippenaar [52] observed that a hydrophobed galena particle with a hor-
izontal orientation (Fig. 16a) had little effect on foam film stability. The
behavior resembled that of spherical particles of contact angle < 90°. How-
ever, in the case of a particle with a diagonal orientation (Fig. 16b) two
consequences arose when it contacted the second air-water surface of the
thinning film. The particle either twisted to a horizontal orientation with two
faces in the planes of the upper and lower film surfaces or retained its di-
agonal orientation. Particles which retained the diagonal orientation caused
rapid film collapse. The sequence of events is illustrated in Fig. 17, where
representations of the original high-speed cinematographic frames are
reproduced.
Here the air-water surface is seen to move up the inclined left-hand face
until the two three-phase contact lines become coincident on the edge,
whereupon film collapse occurs. We therefore find that a cubic particle can
give rise to film collapse, even though 0Aw < 90°, by a mechanism exactly
analogous to that shown for spheres in Fig. 12 and 14.

C. Particle Geometry and Contact Angle Conditions


for Foam Film Collapse
1. Smooth Particles with Edges
It is possible to use the results of Dippenaar [52] to devise general rules
concerning the role of particle geometry in determining film collapse. Thus,
for particles with smooth curved surfaces, such as spheres, cylinders, and
ellipsoids, the condition for film collapse by the process shown in Fig. 12
and 14 is
(27)
where 0Aw is the receding contact angle at the air-water surface measured
through the aqueous phase.
For smooth particles with edges the contact angle for rupture depends
upon the geometry. The basic principles, made obvious by Dippenaar's work
[52], are most clearly revealed if an axially symmetrical particle with three
edges is considered. The geometry of this particle is shown in Fig. 18, whence
it is seen that only one angle 0P need be specified to define all edge angles
(where ep < 90°). If the diameter of the particle is large enough, it will float
at an air-water surface with one of three possible configurations, depending
upon the magnitude of 0Aw· These are illustrated in Fig. 18. In each case
the rotational symmetry axis of the particle will be perpendicular to the air-
water surface.
42 Garrett

Time; ms

12.0

14.9

16.3

17.7

19.1

19.8

--108
----- 97

FIG. 17 Reproduction of high-speed cinematographic film frame showing inter-


action of xanthenated galena particle (OAw = 80 :t 8°) with aqueous film. (From
Ref. 52, reproduced by permission of Elsevier Science Publishers.)

Configuration A corresponds to the condition


0 <()AW< ()p (28)

configuration B to
Op < 0Aw < 180° - !.)p (29)
--
;:::
c.
0
<i-' CD

..
I 0

-+ 18~ ~
)>

--
n

:I
0
stabilising )>
(A) 0 <8Aw <8p :I
:::;;
0
Ill
3


////.
fll

-+IA/~~-+
/TF~

rupturing
(B) 8p < 8Aw < 180° - 8p

-
(C) 180° - 8p < 8AW < 180°
-+
FIG. 18 Effect of axially symmetrical particle with edges on foam film stability.
rupturing (depending on
critical thickness; see
section NC)

J:>
c.>
44 Garrett

and configuration C to
180° - ep < eAw < 180° (30)
Since eP < 90°, film rupture will not occur for a particle with the con-
figuration A when bridging across a foam film occurs. This particle will
interact with the film to produce a capillary pressure to oppose drainage.
On the other hand, particles with configuration B will rupture films accord-
ing to the process depicted in Fig. 18. Thus, condition 29 implies that par-
ticles with the selected geometry may rupture films when eAW < 90°. Par-
ticles with configuration C will only affect film behavior if the thickness for
spontaneous rupture of the aqueous film on the solid is greater than that for
the aqueous film alone. Such a configuration is only possible if eP is in the
region of about 70°-90° because values of 0Aw never usually exceed about
110°.
Particles which exhibit both smooth surfaces and edges will often be crys-
talline with straight edges. These are exemplified by the cubic crystals of
galena used by Dippenaar [52]. This type of particle may be modeled by
projecting the cross section shown in Fig. 18 in a direction perpendicular to
the rotational axis. It is shown in Fig. 19. If the projection distance is long
enough, then configurations in which the particle lies with that projection
parallel to the air-water surface are energetically favored. However, by con-
trast with axially symmetrical particles, coincidence of both three-phase con-
tact lines on an edge when such particles bridge films cannot cause complete
dewetting of the particle. Growth of a hole in the foam film will occur from
the edge and will not be axially symmetrical. That this will occur is of course
confirmed by Dippenaar's observations with galena.
The number of possible configurations at the air-water surface of a straight-
edged particle of given cross section can be much larger than that of an
axially symmetric particle of the same cross section. For example, in the
case of straight-edged particles of cross section defined by one angle 0P shown
in Fig. 19 the number of distinct configurations is 6 when 0P = 30° and the
contact angle 0Aw < 90°. These are shown in Table l, whence it is clear
that not all of these configurations are mutually exclusive for a given contact
angle. Most of them are "degenerate" in that the same configuration can be
realized by using different edges and surfaces. Only one of the configura-
tions may give rise to film collapse.
We may estimate the works of emergence W; of the particle from the
aqueous phase to form each of the possible configurations i shown in Table
1. Here W; is given by
(31)

where 'Ysw. 'YsA> and 'YAw are respectively the solid-water, solid-air, and air-
Mode of Action of Antifoams 45

x- - ---1

\
y
- - ~i:_ j - -

FIG. 19 Hypothetical particle with straight edges.

water surface tensions, and M 5 w, MsA• and MAw are the changes in solid-
water, solid-air, and air-water surface areas accompanying emergence of the
particle into the air-water surface. Since MsA = Msw and we have the
Young equation
'YSA = 'Ysw + 'YAW COS 61\W (32)
then from Eq. 31 we can deduce
W; = 'YAw(LlAsA cos 0Aw - LlAAw) (33)
It is a matter of elementary geometry to estimate Msw and MAw for the
configurations in Table 1. By way of example, values of Wi/'YAwX2 , where
X is the cross-sectional dimension defined in Fig. 19, for each of these con-
figurations are given in Table 1. These are calculated for 0Aw = 40° and
particles of length 5X. Here we see that the only configuration which can
give rise to film collapse has the largest negative work of emergence. How-
ever, this is not the only configuration with a negative work of emergence,
and it has the lowest degeneracy. Clearly, the effectiveness of the particle
in causing film collapse requires consideration of the relative probabilities
of achieving all the possible configurations.
There is no obvious rigorous approach which we may employ to calculate
these probabilities. The problem does, however, resemble that addressed by
classical statistical mechanics where the distribution over energy states of a
system subject to thermal motion is considered. Thus, in general we have
various states available to the particles ranging from free dispersion to adhe-
46 Garrett

TABLE 1 Air-Water-Air Foam Film Collapse and the Configurations at the Air-

Possible configurations at air- Conditions for


water surface if 0Aw < 90° realization of
configurations "Degeneracy"

i = 0 < 0Aw < 30° 2


air

water

i = 2 0 < 0Aw < 30° 4

j = 3 60° < 0Aw < 90° 2

i = 4 15° < 0Aw < 45° 4

i = 5 30° < 0Aw < 150°

i = 6 0 < 0Aw < 90° 2

Particle geometry as shown in Fig. 19 with OP = 30° and Y = X. Z is the length of the particle
defined in the figure.
Mode of Action of Antifoams 47

Water Surface of a Particle with Several Edges

Foam film
collapse if WJ"YAwX 2 for Z = 5X
OAw < 90° Work of emergence W; and OAw = 40°

no WJYAw ZX = (cos OAw - 1)

no WJ"YAw ZX = (cos OAw - 1)/2 sin Op

no WJ"YAw ZX = (cos OAw/sinOP - 1)

no WJ"YAw ZX = (l/(2 sin Op) + l)cos OAw - -2.00


[(1/(4 sin 2 Op)) + l - cos (180 - Op)/sin
OP]1;2

yes WJ"YAwZX = (1 + l/sinOp)cos OAw - (l + -2.17


l /tan Or)

no WJ"YAwX 2 -0.44
= (l + 1/(2 tan OP)) (cos OAw - 1)
48 Garrett

sion at the air-water surface with various configurations. However, during


foam generation agitation will mean that the kinetic energies of the particles
will in general far exceed those due to thermal motion alone. We can attempt
to allow for this by according to the particles a "granular temperature" t in
a manner analogous to that described by, for example, Hopkins and Wood-
cock [68]. Thus, we may write
2ET
T=---- (34)
3(nT - l)k

where nT is the total number of particles, k is the Boltzmann constant, and


3 (nT - 1) is the number of degrees of freedom. Here ET is the total kinetic
energy of the particles so that
l nr
ET= -
2
L m, v~
r=I
(35)

where m, and v, are respectively the velocity and mass of particle r. There-
fore, t is characteristic of the agitation conditions prevailing during foam
generation. If t can be accorded the same significance as temperature in the
relevant context, then we can write a Boltzmann-type expression for the
number of particles n; with configuration i:
n; a exp(-W;l3) (36)
where 13 = 1/kf. In the absence of agitation t =, T, so 13 = 1/kT.
The relative probability q; of any configuration i occurring where the par-
ticle adheres to the air-water surface is therefore given by the familiar Boltz-
mann distribution
g; exp(-W;l3)
q; = s (37)
L g; exp( - W;/3)
i=l

where g; is the "degeneracy" of configuration i and s is the number of pos-


sible configurations at the air-water surface. Low negative or even positive
values for W;, found for the configurations which do not lead to film collapse
shown in Table 1, therefore imply low relative probabilities of occurrence.
The significance of 13 is perhaps easier to see if the relative proportion
of configuration i = 5 (which will cause film collapse) and configuration i
= 4, shown in Table 1, are compared. Thus, we have
(38)

The dimensions of typical antifoam particles are such that X (see Fig. 19)
Mode of Action of Antifoams 49

is in the region of 1-10 µm, so that W 5 - W4 is in the region of -(5 x


10- 13 _5 x 10- 15 ) J (for 'YAw ~ 30 mN m- 1). In the case of thermal motion
alone f3 = 1/kT ~ 2.5 x 1020 r' at ambient temperatures, and we obtain
n4 /n 5 ~ 0, so that the probability of finding configurations of the particles
at the air-water surface which do not cause foam film collapse is essentially
zero. However, where the agitation is intense, it is reasonable to suppose
that the value of f3 could be considerably in excess of 1/kT, so that config-
urations which do not cause foam collapse may become more probable. In-
creasing n4 /n 5 to, for example, 0.01 would require values of l /f3 from (2
x 105-2 x 107)kT, depending on the size of the particle.
In principle it is possible to deduce similar conditions to those depicted
in Table 1 for any crystal habit. The resulting conditions may then be com-
bined with contact angle measurements to deduce the antifoam potential of
derived crystalline particles. Apart from the work of Dippenaar [52], con-
cerning galena, no studies of the antifoam behavior of such well-defined
hydrophobic crystalline particles have been made. However, Frye and Berg
[63] have demonstrated the effect of sharp edges by comparison of the an-
tifoam effectiveness of polydimethylsiloxane hydrophobed smooth spherical
and ground-glass particles. Here ground glass formed shards with relatively
few edges. Contact angles were adjusted by using different concentrations
of several surfactants. The relative effects of the different types of particle
are shown in Fig. 20. The antifoam effectiveness of ground-glass particles
for a given contact angle is clearly much greater than that of smooth spheres.
That there is, however, an apparent indication of antifoam effects for
smooth spheres with contact angles less than 90° is attributed by Frye and
Berg [63] to the presence of a minor proportion of nonspherical particles
with sharp edges among the spheres. These authors also found these hydro-
phobic ground-glass particles to be more effective than PTFE particles pre-
sumably because of the greater incidence of sharp edges with the former.
2. Rough Particles with Many Edges
Sparingly soluble materials are often found precipitated from aqueous so-
lution in the form of amorphous particles of ill-defined structure. Precipi-
tated silica is an obvious extreme example where fractal structures are formed
(of dimension about 2.4) [69]. Such materials may be intrinsically hydro-
phobic or be rendered hydrophobic by surface treatment.
In contrast to the ground-glass particles used by Frye and Berg [63], ran-
dom rough amorphous and hydrophobic particles may in general adopt many
different possible configurations at the air-water surface for a given contact
angle. Some of these may be associated with film collapse if the requisite
contact angle conditions are satisfied. However, the existence of many edges
and asperities will increase the chances of a planar interface hinging at more
50 Garrett

Foam
removal
(vol %)

100
• • •• • ••c.. ~<>
• •• CU
• •
80 0
<>
60
D 0
0
40 0 0 <>
20
oD
0 0 <>
25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Contact angle 8Aw I degrees

I Ground glass; II sodium dodecyl benzene sulphonate, e Triton X-100, • hexadecyltrimethyl


ammonium bromide
Spherical glass particles; o sodium dodecyl benzene sulphonate, O Triton X- 100,

<> hexadecyl trimethyl ammonium bromide

FIG. 20 Antifoam effect of hydrophobed glass particles. (From Ref. 63, repro-
duced by permission of Academic Press Inc.)

than one site on a particle. Stable bridging configurations in aqueous foam


films may then occur with two planar air-water surfaces. The case of an
axially symmetrical rough particle with regular rugosities is shown in Fig.
21 by way of example. Here only one of the allowed configurations which
satisfy the contact angle eAw condition

0, < e < 180° -


AW 0, (39)

(where e, < 90° and


is defined in Fig. 21) can give rise to film collapse.
For that configuration the air-water surface is depicted in Fig. 21 as hinging
on rugosity ii = 8 so that most of the particle is outside the aqueous phase.
All other configurations would actually stabilize a foam film against drain-
age. Moreover, the probability of occurrence of the destabilizing configu-
ration is relatively low if 0Aw < 90°. This is revealed by consideration of
the relative values of the work w:
of formation of each configuration. Here
wt may be calculated in a similar manner to W; by using Eq. 33.
Mode of Action of Antifoams 51

cp
I

ii= 1
ii= 15
ii=2
ii= 14
ii=3
ii= 13
ii=4
ii= 12 Particle will adhere to the
ii=5 air-water surface at any
ii= 11 rugosity ii if
ii=6
ii= 10
e, <8Aw < 180° - e,
ii=7
ii= 9
ii=8

x -----

-0-_..lJ=
(a)
air

.. ~r
air

L _. film rupture

FIG. 21 Axially symmetrical particle with many edges: (a) example of configu-
ration which stabilizes foam film, (b) configuration with air-water surface hinging
on rugosity ii = 8 will rupture foam film.
52 Garrett

TABLE 2 Work of Emergence into Air-Water Surface, W;~ of Axially Sym-


metrical Particle with Regular Rugosities

Configuration with Wtf'YAwX2


air-water surface
hinging on
rugosity ii (lAW = 45° 0AW = 85° (lAW = 95°
ii =1 -0.347 -1.291 -1.557
2 0.695 -1.163 -1.685
3 1.738 -1.034 --1.814
4 2.781 -0.905 -1.943
5 3.823 -0.777 -2.071
6 4.866 -0.648 -2.172
7 5.908 -0.520 -2.328
8 6.951 -0.391 -2.457
With 0, = 30° and H/X = 0.1; see Fig. 21 for definition of 0,. H, and X.

The geometry of the particle shown in Fig. 21 means that we can write,
for w;,
w;
- -72
-
'YAwX
= 1T [Pii (H/X) ( + -H/X)
-.--
sm 0,
- + I
tan 0,
]
0.25 cos eAw
(40)

[0.5 + -
H/X] 2
- 1T -
tan er

where X is defined in Fig. 21 and where Pu is the number of dewetted conical


segments each of height H associated with the air-water surface hinging on
rugosity ii. Values of w;
hAwX2 for different configurations of the particle
are presented in Table 2 for three different contact angles by way of example.
Low negative values or even positive values for W;~ found for the desta-
bilizing configuration ii = 8 shown in Table 2 for 0Aw < 90° imply low
relative probabilities of occurrence. Conversely, for 0Aw > 90° the desta-
bilizing configuration has the highest negative value for which implies w;,
the highest probability of occurrence.
In conclusion, it is tempting to generalize from this example that for rough
particles with many edges and asperities the contact angle for destabilization
cannot be specified with certainty so that antifoam effects will even occur
with 0Aw < 90° (provided condition 39 is satisfied). However, destabilizing
configurations for 0Aw < 90° will occur only with low probability, so that
the overall antifoam behavior will be relatively weak or even negligible in
those circumstances. Thus, for film collapse the particle must adopt a con-
Mode of Action of Antlfoams 53

figuration at the air-water surface so that no rugosities exist below that sur-
face at which the second air-water surface of a foam film could hinge. Such
configurations of necessity involve removal of most of a particle with many
rugosities outside the aqueous phase. For contact angles < 90° this is en-
ergetically unfavorable and will therefore have a relatively low probability
of occurrence.

D. Particle Size and Kinetics


An essential feature of the bridging mechanism for antifoam action by par-
ticles is that the particle size be of the same order as the thickness of the
foam film. Foam films drain and stretch during foam generation and after
foam generation has ceased. The rate of such processes will clearly play an
important role in determining the frequency of foam film collapse if the film
must first drain to the dimensions of any particles present.
Dippenaar [52] has developed a simple model to describe foam film col-
lapse as a function of particle size for the case of foam generation by shaking
cylinders. With this method of foam generation repeated shaking will often
eventually produce a constant foam volume. Here presumably the rate of
production of foam equals the rate of destruction due to the presence of
hydrophobic particles. The rate of destruction according to Dippenaar [52]
will, however, be proportional to the amount of foam present. Therefore the
final foam volume is proportional to the volume rate of foam destruction.
That volume may be maintained constant for different particle sizes by ad-
justing the total mass M of particles present. Then M becomes the mass of
particles required to reduce the final foam volume to the chosen reference
value. All of this implies a relationship between particle size and the total
mass of particles required to achieve a constant volume rate of foam de-
struction (which is proportional to the rate of film rupture).
At the supposedly constant final foam volume the frequency fr with which
a single particle ruptures a film is simply the rate of film destruction/total
number of particles. Therefore fr is given by [52]

(41)

where a is the particle diameter and k 1 is a constant.


As we have seen, for particles to rupture thin films they must first thin
down to a thickness proportional to the particle dimension (dependent upon
particle size, shape, and contact angle). The time taken for this to occur is
difficult to assess with certainty. Thus, film drainage is by viscous flow
between the air-water surfaces of the film under the influence of gravity or
54 Garrett

capillary suction at the plateau borders. It may also possibly occur by mar-
ginal regeneration (70]. The relative significance of these processes depends
upon film orientation and size. However, in the case of large (>a few square
centimetres) vertical free nonrigid (i.e., "mobile") films Mysels et al. [70]
show that marginal regeneration makes the most significant contribution to
film drainage.
The time t taken for a ve11ical free rigid liquid film to thin to the thickness
h by Poiseuille flow under gravity is [70]

41]y
t=-- (42)
pgh2

where lJ is the viscosity, y is the height above the bulk liquid surface, and
g is the acceleration due to gravity. In the case of a horizontal rigid plane-
parallel circular film subject only to capillary suction from the plateau border
the time taken may be deduced from the Reynolds equation [71], so that

3TJR 2
t=--- (43)
2 M'rh2

where R is the radius of the film and !:.Pr is the capillary pressure. Marginal
regeneration, on the other hand, is a complex process, and no expression is
available for t as a function of h.
In the derivation of both Eqs. 42 and 43 rigidity not only implies absence
of marginal regeneration but also that the velocity of flow at the surfactant
contaminated air-water surface is zero. Violation of the latter assumption,
even supposing absence of marginal regeneration, would mean shorter film
drainage times than those predicted using these equations.
Dippenaar [52] ignores any contribution from marginal regeneration. He
supposes that the time taken for a film to thin to the dimension required for
rupture by a particle is given by Eq. 43 and is therefore simply inversely
proportional to the square of that dimension. This would be true if drainage
is described by either Eq. 42 or 43, where under the conditions of constant
final foam volume either y or R2 / t::.P r are constant. From Eq. 16 Dippenaar
[52] therefore obtains the relationship

(44)

where k2 is another constant and tr is the time taken for the film to thin to
the thickness where rupture by a particle can occur. Using a range of hy-
drophobic quartz particles differing by more than two orders of magnitude
Mode of Action of Antifoams 55

in size, Dippenaar [52] shows that a/M is essentially constant for foams
generated from aqueous solutions of tricthoxybutane. This implies, if we
take account of Eq. 44, that the frequency of rupture of foam films is in-
versely proportional to the time taken for the films to thin to a thickness
where rupture by a particle can occur. Dippenaar [52] then concludes that
the "rate-determining step" for film collapse by particles is "natural" thin-
ning of films. This result is, however, obtained by using an expression for
that natural thinning process which is only true if marginal regeneration is
absent. It seems unlikely that aqueous solutions of triethoxybutane will form
rigid films. It is, however, possible that marginal regeneration makes little
contribution to film thinning (even for solutions which form mobile films)
when the films are of the small size of those formed between bubbles in
foam freshly generated by cylinder shaking. Marginal regeneration appears
to be an oscillatory phenomenon, which may mean that if the dimensions
of the film are less than a certain critical wavelength then it does not occur.
t;
Equation 44 also means that if j~ is proportional to 1 then for a constant
particle size the ratio TJ/M must be constant. Dippenaar [52] has verified
this, using glycerol to modify solution viscosity. It implies that the mass of
antifoam required to achieve a given final volume of foam is proportional
to the viscosity of the solution.
If the rate-determining step for film rupture is generally the time taken
for films to thin, then the actual process of film rupture, once particle bridg-
ing has occurred, must be relatively rapid. Frye and Berg [63] have devel-
oped a model of that process which yields estimates of the rupture time.
The model assumes an axially symmetrical configuration similar to that shown
in Fig. 22. Here the air-water surface in the vicinity of the particle is as-
sumed to have a circular cross section between the contact line on the par-
ticle and flat surface of the plane parallel film. The capillary pressure APc
in the vicinity of the particle is crudely simplified by considering only the
curvature in the plane of Fig. 22 (i.e., Eq. 26). The rate of flow of fluid Q
from the particle under the influence of APc is then calculated under the
assumption that all the displaced liquid is squeezed out of the film so that
it remains plane-parallel except in the vicinity of the particle. A rigid circular
film is assumed, so that Q is given by
APch 3
Q= (45)
'lT

3TJ
This expression is readily obtained by differentiating Eq. 43, replacing APr
by Mc, and noting that Q = '1TR 2 dh/dt.
The position of the three-phase contact line is adjusted by small incre-
ments fJ.h* from h* = 0 to h* = h/2, where h is the thickness of the film
(see Fig. 22). Here h* is the distance from the plane of the film surface to
56 Garrett

C!.-'
I

_.,' -

volume fluid
• displaced as
contact line moves

FIG. 22 Illustration of procedure for estimation of time of antifoam action. (From


Ref. 63, reproduced with permission of Academic Press Inc.)

the plane of the three-phase contact line. The time increment corresponding
to each increment 11h* is calculated from the relationship

11t=l1V (46)
Q
where /1 V is the corresponding incremental change in volume of fluid re-
moved from the shaded area in Fig. 22. The time to rupture is then N 11t,
where the number of increments N is given by N 11h* = h/2.
The assumption that all the displaced fluid is completely removed from
the film so that it remains plane parallel facilitates calculation but may not
be entirely realistic. In practice, liquid may migrate only a small distance
from the particle to form a bulge. Calculated rupture times may therefore
be overestimates.
Rupture times calculated with these procedures are dependent upon con-
tact angle and particle size. Thus, as the contact angle declines to the ap-
propriate critical value, the rupture time asymptotically approaches infinity.
For example, rupture times <10- 3 s require contact angles about 15° higher
Mode of Action of Antifoams 57

than the critical value of 90° for a rod-shaped particle of 0.1 cm radius.
Some experimental evidence is presented by Frye and Berg [63], which in-
dicates that effective antifoam action requires contact angles significantly
higher than the critical values due to the effect of rupture time. Thus, in
experiments with hydrophobic rods of 0.1 cm radius and free foam films,
they show that the probability of perceived instantaneous rupture is signif-
icantly less than unity for advancing angles of > 100° where the critical angle
for the rod is 90°. However, the relevant data are badly scattered. Moreover,
the experiments are ambiguous with respect to comparison with measured
contact angles. Thus, an advancing angle occurs at the upper surface of the
film, and a receding angle should occur at the lower surface. The experi-
ments are also ambiguous with respect to the time allowed for rupture. A
perception of instantaneous rupture presumably implies rupture times of <
0.1 s, which would require, according to the calculation of Frye and Berg
[63], a contact angle of only 95° for rods of 0.1 cm radius.
The larger the particle the greater the volume of liquid which must be
squeezed out of the film before rupture can occur. Rupture times therefore
increase with particle size. However, Frye and Berg [63] show that a three-
orders-of-magnitude increase in particle size for rod-shaped particles results
in a difference of only about I 0° in the required contact angle if film rupture
is always to occur in less than -10- 3 s.
Frye and Berg [63] also use Eq. 45, with llFr instead of APc, to calculate
the thinning time which they compare with the rupture time to estimate the
rate-determining step. For small particles (<if 100 µm) rupture times are min-
imal and thinning is rate determining in agreement with the conclusion of
Dippenaar [52]. However, if the particles are sufficiently large (;?;> 100 µm),
then thinning times decline and rupture becomes rate determining. Frye and
Berg [63] calculate an optimum particle size for which the total time of
antifoam action is minimal. Rates are dominated by rupture times for sizes
larger than the optimum and by thinning times for sizes smaller than the
optimum. We should note here, however, that particles of optimum size for
minimal antifoam action time are not necessarily of optimum size for overall
effectiveness. A dominating factor is that the smaller the particle size the
greater the number of particles for a given antifoam mass and the higher the
probability of film rupture. This is shown empirically by Dippenaar
[52], where as we have seen the mass M of antifoam required to reduce
the foam volume to a chosen reference value is proportional to the particle
size.
Frye and Berg [63] provide no detailed comparison with experiment for
these considerations. However, they note that the deviation from constant
a/M recorded by Dippenaar [52) for extremely large particles is consistent
with rupture times becoming rate determining.
58 Garrett

These calculations concerning total time for antifoam action rely essen-
tially upon the relevance of the Reynolds equation [71] and require knowl-
edge of Mt which is not readily accessible by experiment for a real foam.
Moreover, it is possible that during foam generation the velocity of flow at
the air-water surfaces of films is nonzero so that film-thinning rates are faster
than predicted by the Reynolds equation. The calculations also start with a
configuration where the particle is already dewetted into one side of the foam
film, and implicity assume that the formation of a second three-phase contact
line is instantaneous once the film thins to the relevant thickness. The pro-
cess by which particles dewet into the air-water surface is ignored. We now
address this issue.

E. Capture of Particles by Bubbles


Formation of a three-phase contact line is of course associated with rupture
of the aqueous film separating the particle from the air bubble. The process
of rupture of that film is usually believed to occur by catastrophic growth
of random fluctuations in the air-water surface of the film at a certain critical
film thickness hcrit in a manner analogous to that described by Vrij and
Overbeek [9,72] for free aqueous foam films (see Sec. II.B.). Here we re-
member two factors influence the growth of those fluctuations. The first is
the work done against the air-water surface tension in perturbing the surface,
and the second is the work of interaction due to the disjoining force acting
across the film.
In the case of hydrophobic particles interacting with air bubbles in an
aqueous surfactant solution, both attractive and repulsive disjoining forces
will exist. The latter will often be electrostatic in origin and are of relatively
long range. They may be due to overlapping electrostatic double layers caused
by adsorbed charged surfactant. They may even be a consequence of charge
intrinsic to the solid surface in an aqueous environment. Thus, strong elec-
trostatic repulsive forces stabilize films of pure water on silica even when
the latter is hydrophobed with methyl silanes (73].
Van der Waals attractive forces tend to be relatively short range. The
resultant negative work of interaction exhibits a distance dependence of h- 2 ,
which obviously becomes large as h-'-"O [74). In contrast, the positive work
of interaction between overlapping electrical double layers often approxi-
mates a distance dependence of exp(-Kh) (where K is the Debye length),
which tends to a constant value as h-'-"O [7 4 j. For hydrophobic solids the
van der W aals contribution therefore tends to dominate as h-'.>0 leading to
negative values of the work of interaction. However, at intermediate dis-
tances the two types of interaction often conspire to produce an energy max-
imum at h = hmax 0
Mode of Action of Antifoams 59

The growth of fluctuations is resisted by the need to do work against the


air-water surface tension when perturbing the film (because the surface area
increases). If a catastrophic growth in fluctuations is to occur, then this re-
sistance must be overcome by the tendency of the film to thin spontaneously
due to the dominance of attractive disjoining interactions.
A complete analysis of the hydrodynamic stability of a thin film between
a hydrophobic solid surface and an air-water surface where the interaction
potential exhibits a maximum at hmax does not appear to be available. It
seems probable, however, that in this case hcrit < hmax· If a particle is to
adhere to an air bubble, it must first have sufficient kinetic energy to over-
come this energy barrier and sufficient time for the film between the particle
and the bubble to attain the critical thickness and rupture.
Arguably one of the most careful studies of particle-bubble interactions
under dynamic conditions is that due to Anfruns and Kitchener [75]. Here
the efficiency of capture by rising bubbles of glass microspheres and rough
quartz particles, rendered hydrophobic with methyl silanes, was determined
under carefully controlled conditions. Bubbles were in the size range 0.05-
0.1 cm, and particles were of size about 30 µm. These experiments enabled
the collection efficiency Ce to be obtained as

number of particles collected


C=-----=-------
e 1T R'f,Lnp
(47)

where Rb is the radius of the bubble, L is the path length of the bubble, and
nP is the number of particles/unit volume suspended in the medium through
which the bubble passes. Here Ce ~ I because most particles ahead of a
rising bubble are swept along the fluid streamlines remote from the bubble
surface. The process is depicted schematically in Fig. 23.
The conditions of these experiments were such that the relevant stream-
lines could be calculated so that Ce could in tum be calculated. In Figs. 24A
and B the calculated and experimental values of Ce for both smooth glass
microspheres and rough quartz particles are compared for various solution
conditions. The efficiency of collection of the glass microspheres is seen to
be considerably less than that of the quartz particles under all these condi-
tions. Indeed the efficiency of collection of the quartz particles generally
approaches theoretical values. This means that every quartz particle follow-
ing a streamline which takes the particle within a particle radius of the sur-
face of a bubble is captured by the bubble. These differences between the
quartz and glass particles occur despite the similarity of both the surface
treatment and contact angles measured on representative smooth surfaces
(- 85-90° in distilled water). Similar observations concerning the effec-
tiveness of surface roughness in promoting the efficiency of capture of par-
60 Garrett

I: I
I I I. I
I I I
I I
I
I
I I

\/
I I
I I; I I 1 Streamlines

/ ;;+1'
I I
I I
I \
I I
I / I I \ I I
I I I j \ \ \
I / / I I \ I
I I / \ , \
.._c--- '-, '
\
I I
'
// \

\~
I I / / '- '- \
1 / f /' '\\ \
1 I I I ' \ I
I I I I 1
A I / 1. I
TI \ I I :
I
\
\
I
I
\
I
\
/

! I
\
'
\ • \ / /:' I
\' ',,
\ \ ~ / / /
/ /
\ \
\ '\ "- I
/ /
/ /
\ '\,,.//I
\ Streamlines
'\ \ \ I I I
\ '- \ \+( I I
within radius of
\ \ \ I .I I I particle (allowing
I
\
\ I '1 : I I I for Brownian
\ I I I : I I I motion) so it can
I I
\ I I; I I touch bubble
\ 'I I
I: surface
I:
I
I I
I
I I
Ii 1
I I
\ I
I

FIG. 23 Bubble rising through suspension of hydrophobic particles under laminar


flow illustrating inefficient interception of particles by bubbles (schematic).

tides by bubbles are reported by Ducker et al. [76] and Strnad et al. (77].
Interactions due to electrical double layers are essentially suppressed by
the addition of 0.1 M KCl. This is seen from Fig. 24B to more than treble
the collection efficiency of glass microspheres. Electrical double layers are
therefore clearly implicated in contributing' to the relatively low collection
efficiency of these particles. Further evidence of the importance of electro-
static interactions is provided by the effect of adding sodium dodecylsulfate
up to 10-3 M. The surfactant adsorbs at the air-water surface to produce an
increase in electrostatic repulsion between the particles and the bubbles (75].
This virtually eliminates capture of glass particles despite contact angles of
30° prevailing in these circumstances.
Anfruns and Kitchener [75) deduce that relatively long-range electrostatic
repulsion forces are preventing the effective thinning of the aqueous film
Mode of Action of Antifoams 61

- - - Theory

Collection O distilled water


Efficiency
• 1Q-4M sodium dodecyl sulphate
0.04
o 10- 3 M sodium dodecyl sulphate

0.03
A
j 1M KCL

0.02

0.01

0.5 0.6 0.7 0.8 0.9 1.0


0.04

0.03
B
0.02 A A 11.

O.Qi
0
A A
,, '•l A

0.5 0.6 0.7 0.8


~t]J
0.9
'
1.0

Bubble diameter I mm

FIG. 24 Collection efficiencies for rough quartz and spherical glass particles: (A)
quartz particles of 31 micron Stokes equivalent diameter; (B) spherical glass particles
of 32 micron diameter. (From Ref. 75, reproduced by permission of the Institute of
Mining and Metallurgy, London).
62 Garrett

~
hmax

FIG. 25 Approach of smooth spherical particle and spherical particle with asper-
ities to air-water surface (hmax exaggerated).

between smooth glass particles and bubbles during the time available when
the particle is in near contact with the rising bubble. In contrast, sharp edges
and asperities on the surface of the rough particles facilitate rupture of the
film despite the presence of the long-range electrostatic forces.
Most antifoam particles are either rough or possess edges. This is also
usually true of mineral particles involved in flotation. The effect of edges
on capture by bubbles would therefore appear to be of general importance
in understanding particle-bubble interactions. Unfortunately no proper the-
oretical explanation is apparently available. A naive explanation is, how-
ever, suggested if we compare a spherical particle of radius rP having a
spherical segment asperity of radius r" with a smooth spherical particle of
the same size. As we have seen, dewctting probably requires these particles
to have sufficient kinetic energy to overcome a repulsive energy barrier.
Mode of Action of Antifoams 63

Now imagine the two particles to approach the air-water surface of a bubble
in the manner shown in Fig. 25, where deformation of the air-water surface
is ignored to a first approximation. The ratio of the height of the energy
barrier for the asperity to that for the smooth particle as a whole is r./rp,
where r. < rP and provided r.. ~ hmax (so that we may assume Derjaguin's
approximation [78]). Clearly, then, the particle could have sufficient kinetic
energy to dewet the asperity and initiate attachment to the bubble without
necessarily having sufficient energy to do so in the absence of the asperity.
Presumably the asperity could then have a similar effect on the stability of
the film still separating most of the particle from the air bubble to that shown
by particles bridging free foam films.
Frye and Berg [63] in their study of the effect of particle morphology on
antifoam effectiveness were apparently unaware of the work of Anfruns and
Kitchener [75]. Clearly, however, they selected similar systems to those
selected by the latter. The superior antifoam action of hydrophobed ground-
glass particles relative to the effect of spherical glass particles, shown in
Fig. 20, may therefore owe something to the more ready attachment of rough
particles to bubble surfaces. The importance of roughness in determining
dewetting behavior does hint that the contribution to the weak antifoam be-
havior of the spherical particles by a contaminating proportion of particles
with sharp edges is likely to be even more important than Frye and Berg
[63] indicate.

F. Melting of Hydrophobic Particles and Antifoam


Behavior
Certain hydrophobic materials which exhibit antifoam effects have melting
points at temperatures < 100°C. Examples include hydrocarbon waxes, tri-
glycerides, and long-chain fatty acids.
Dispersions of hydrophobic particles can be prepared by first emulsifying
the molten materials and then cooling the emulsions below the melting point.
Observations by Davis and Garrett [79] of the antifoam behavior of the re-
sulting entities reveal some interesting behavior. By way of example in Fig.
26 we plot the ratio F, where
volume of air in foam in presence of antifoam
F= ~~~~~~~~~~~-~~~~~-

vo Ium e of air in foam in absence of antifoam


against temperature for 1.2 g dm- 3 dispersion of n-docosane, n-eicosane,
and paraffin wax in 0.5 g dm - 3 sodium (C 10-C 14 ) alkylbenzene sulfonate
solution. Foam measurements were made in a static Ross Miles apparatus.
Particle sizes were in the range 0.5-8 µm.
The most striking feature of the behavior of n-docosane is the sharp de-
terioration of antifoam effect in the region of the melting point at 44°C.
64 Garrett

F
I n-Docosa~ F
In-Eicosane I

f.- ;
1.0 1.0
Q/"0-
~
0.9 0.9 D
oNPL sample
0.8 0.8
•Aldrich
0.7 sample 0.7 B

),
0.6 0.6
0.5
0.4
0.5
0.4
1\
0.3 0.3
0.2 .... 0.2 A
c
0.1 0.1 ~Melting Range

10 20 30 40 50 60 70 10 20 30 40 50 60 70
Temperature t °C Temperature I °C

F
1.0 Paraffin
Wax
B
0.9
D
0.8
~

\
0.7
0.6
0.5
0.4 0

0.3 A

0.2
\ c
0.1 Melting~
Range

10 20 30 40 50 60 70
Temperature I °C

FIG. 26 Effect of temperature on the antifoam behavior of finely divided hydro-


carbons (1.2 g dm- 3 hydrocarbon in 0.5 g dm- 3 commercial sodium (C10-C14)
alkylbenzene sulfonate (Dobs 055) solution using static Ross Miles). (From Ref.
79.)
Mode of Action of Antifoams 65

An extremely pure sample was used, and therefore a sharp melting tran-
sition was observed. Similar observations have been made by Aronson
[80] for hydrocarbons and fatty acids in solutions of sodium tridecylbenzene-
sulfonate.
Neither the n-eicosane nor the paraffin wax used in this work were pure
materials. Therefore both exhibit a relatively wide temperature range
over which both solid and liquid phases coexist. The antifoam behavior
of these materials, shown in Fig. 26, is therefore complex. At low tem-
peratures in region AB an antifoam effect associated with solid hydrocar-
bon particles is present. Curiously, that effect appears to deteriorate with
increase in temperature. However, in region BC, where both solid and
liquid hydrocarbon coexist, a pronounced enhancement of the antifoam
effect is seen to occur. This type of behavior has apparently also been
observed by Kulkarni et al. [81] for impure triglycerides. At sufficiently
high temperatures where all solid is converted to liquid the antifoam ef-
fect is seen to deteriorate sharply in much the same manner as for pure
n-docosane.
That antifoam effects may be associated with solid wax or fatty acid par-
ticles is not altogether surprising. Under microscopic observation the par-
ticles are observed to exhibit more irregular shapes than the emulsion spheres
from which they are derived. Aronson [80] observes particularly marked
roughness with many sharp edges for particles of stearic and palmitic acid.
These particles are also found to be more effective antifoams than the more
symmetrical hydrocarbon particles, for which similar contact angles would
be expected to prevail. The equilibrium receding contact angles for the hy-
drocarbons used to obtain the results shown in Fig. 26 are about 105° against
distilled water, which decrease to about 30° against the relevant alkylben-
zene sulfonate solutions [79]. It seems probable then that a combination of
edges/asperties, finite contact angles, and dynamic effects (where the dy-
namic contact angle is higher than the equilibrium contact angle due to slow
surfactant transport to the relevant surfaces) will mean that antifoam behav-
ior by a bridging mechanism will occur. Both Davis and Garrett [79] and
Aronson [80] present evidence which suggests that dynamic effects are im-
portant for these systems.
Gradual deterioration of antifoam effectiveness of the solid particles with
increase in temperature in region AB for the impure hydrocarbons in Fig.
26 is difficult to explain. No significant change occurs in contact angles over
the relevant temperature range [79]. It is, however, possible that the hydro-
carbon softens with increasing temperature, so that asperities and edges are
gradually removed.
We address the finding that liquid hydrophobic oils are almost completely
66 Garrett

F
1.0

0.8

0.6

0.4

• •
• •
0.2

/I
5 10 15 20 25 / 95 100
hydrophobed silica I weight %

FIG. 27 Antifoam effectiveness F for dispersion of hydrophobed silica/mineral oil


antifoam as a function of anti foam composition (silica D 17 ex Degussa; 1. 2 g dm - 3
antifoam in 0.5 g dm 3 commercial sodium (C 10-C 14 ) alkylbenzene sulfonate (Dobs
055) solution; foam generated by cylinder shaking at ambient temperature 22 ±
2°C). (From Ref. 40.)

ineffective as antifoams and that mixtures of hydrophobic liquids and solids


are particularly effective in the next section.

V. MIXTURES OF HYDROPHOBIC PARTICLES AND


OILS AS ANTIFOAMS FOR AQUEOUS SYSTEMS
A. Antifoam Synergy
One of the most striking aspects of antifoam behavior is the synergy shown
by mixtures of insoluble hydrophobic particles and hydrophobic oils when
dispersed in aqueous media. We illustrate this behavior in Fig. 27 with a
plot of the ratio F against antifoam composition for a mixture of methyl-
silane hydrophobed silica and a mineral oil. Foam from a solution of 0.5 g
dm - 3 ( ~ 1.4 x I 0- 3 M) commercial sodium (C 10-C 14) alkylbenzene sul-
fonate was generated by cylinder shaking [40]. The mineral oil is seen to
be virtually without effect on foam behavior, and the silica particles exhibit
only a weak effect. Adding only a few percent of hydrophobed silica to the
mineral oil is, however, enough to produce a significant enhancement of
Mode of Action of Antifoams 67

antifoam performance. Similar observations are reported by others for mix-


tures of hydrophobed glass with hydrocarbon [39] and hydrophobed silica
with polydimethyl siloxanes [30,81-84]. With the latter, silica may be first
hydrophobed by use of a suitable agent, such as a silane or alcohol, before
mixing with the polydimethylsiloxane. Alternatively the silica may be hy-
drophobed in situ by heating with polydimethylsiloxane to temperatures in
excess of l 50°C (possibly in the presence of a catalyst) [85]. The nature of
the reaction between silica and polydimethylsiloxane is not well understood.
Antifoams produced by this process are often described as silicone
"compounds."
We have found that this antifoam synergy appears to be quite general for
all manner of hydrophobic particles and oils [40]. Thus, intrinsically hy-
drophobic organic particulates, such as precipitates of polyvalent metals with
long-chain alkyl phosphates and carboxylates, may also be combined with
mineral oils [86,87] or polydimethylsiloxanes [88] to produce synergistic
antifoam behavior.
Mechanistic studies of this phenomenon did not appear in the scientific
literature until the mid-seventies [30,81,83 ,84]. Indeed only a few refer-
ences to it appeared before 1970 [29,89]. However, descriptions of many
examples of mixtures of hydrophobic particles and oils have been appearing
in the patent literature since the early fifties. Examples of those patents are
listed in Tables 3-5. Here they are categorized as mixtures of hydrophobed
mineral particles and silicone oils (Table 3), mixtures of hydrophobed min-
eral particles and organic liquids (Table 4), and mixtures of intrinsically
hydrophobic organic particles and organic liquids (Table 5).
The finely divided nature of the particles is often stressed in these patents
and methods of milling are sometimes described [ 108]. Whenever particle
size is quoted, it usually falls in the range 0.001-1.0 µm. The preferred
particle concentration in the oil ranges from I% to about 30% by weight.
As we have seen, the antifoam performance of both hydrocarbons and
polydimethylsiloxane oils may be considerably enhanced by the addition of
finely divided silica. This suggests an obvious phenomenological similarity.
However, when hydrophobed silica is added to polydimethylsiloxanes, it is
often referred to as a "filler" 183,92,94,95] or "activator" [93J, implying
that the oil plays the active role in the antifoam mechanism. On the other
hand, when hydrophobed silica is added to hydrocarbon oils, the oil is some-
times referred to as the "carrier," implying that the silica plays the active
role in the antifoam mechanism. Similarly when long-chain organic mate-
rials are added to hydrocarbons, the particles are often referred to as the
"active" ingredients and the oils as the "carriers" (see, for example, ll08]),
which once again implies that the particles play the active role in the anti-
foam mechanism.
TABLE 3 Some Examples of Antifoam Patents Concerning Mixtures of
Hydrophobed Mineral Particles and Silicone Oils

Preferred
Actual concentration
Oils claimed to Particles claimed examples of particles/
be effective to be effective given % by weight

Partially oxidized Silica aerogel pre- As claimed. 7.5


methyl siloxane sumably rendered
polymer. May be hydrophobic by re-
diluted with di- action in situ with
methyl siloxane the siloxane.
polymer and dis-
persed in benzene.
Methyl Finely divided sil- Silica + methyl 2-10
polysiloxane. ica presumably ren- polysiloxane +
dered hydrophobic emulsifiers.
by reaction in situ
with the siloxane.

Silicone oils (prod- Aluminium oxides, 1. Dimethyl polysi- 1-30


uct may be diluted titanium dioxides, loxane + silica +
by hydrocarbons, particularly all aerogel + AlC'3 di-
ethers, ketones or manner of silicas. luted by toluene.
chlorohydrocar- These react with 2. Dimethyl polysi-
bons. This does silicone oil in situ loxane + alumina
not, however, ap- catalyzed by an + SnC14 diluted by
pear as a specific acid condensation toluene.
claim). reagent. 3. Dimethyl polysi-
loxane + precipi-
tated silica + phos-
phorous nitrile
chloride + polyeth-
ylene glycol
stearate.
Silicone oils. Pyrogenic or pre- As claimed. 2-8
cipitated silicas hy- (preferably 5)
drophobed with
"chemically bound
methyl groups"*.
Special reference is
made to the relative
ineffectiveness of
the oil alone
(*probably
silanized).

68
Particle Any other
size additives Application Date Ref.
Aqueous alkaline solns. 1953 90
for paper industry, rub-
ber industry, metal-
working industry.

"Finely Patent really concerns Patent particularly con- 1958 91


divided". emulsifiers which are cems emulsifiers for sil-
monostearic acid ester icone antifoams for cos-
of polyethylene glycol metics and drug
+ monostearic acid es- application.
ter of sorbitol.
0.01-25 l . Acid condensation Alkaline aqueous solns. 1966 92
micron. catalyst necessary (e.g., particularly alkaline
AICl 3 , SnC1 4 BF3 , etc.) cleaning products for
2. Emulsifiers such as automatic washing
methyl cellulose, poly- machines.
ethylene glycol mono-
stearate, polyethylene
glycol trimethyl nonyl
ether added to prepare
stable aqueous 0 /W
emulsions of the anti-
foam if desired.

0.015-0.05 Aqueous surfactant 1968 93


micron. solutions.

69
70 Garrett

TABLE 3 continued

Preferred
Actual concentration
Oils claimed to Particles claimed examples of particles/
be effective to be effective given % by weight

Organosiloxane liq- Hydroxyl contain- 1. Dialkyl amino 1-30


uid polymers which ing inorganic "fill- silicone treated sil-
may be partly or ers" Ti0 2 , Al 203 ica in dimethyl sil-
entirely replaced by and preferably Si0 2 oxane polymer di-
other nonaqueous reacted with a di- luted with
fluids such hydro- alkyl-amino-orga- polypropylene
carbon oils, poly- nosilicone. Special glycol.
alkylene glycols provision of patent 2. Dialkyl amino
etc. is that necessity for organo silicone-
catalysts is ob- treated silica in
viated. Heating and polypropylene
cooling cycles are glycol.
also obviated. 3. Dialkyl amino
organosilicone
treated silica in
mineral seal oil di-
luted with polypro-
pylene glycol.
Dimethyl polysilox- Silica of surface Silica + polydime- l-10
ane or about equal area >50 m 2 gm I thy! siloxane.
proportions of mix- (List of previous
ture of dimethylpo- silicone antifoam
lysiloxane + dior- patents given as ex-
ganopolysiloxane amples of the type
(containing silicon- of materials
bonded Me, Et, considered.)
and 2-phenylpropyl
groups).
Alkylated polysi- Silicas rendered hy- Dimethyl siloxane I0-50
loxane materials of drophobic by a va- + trimethyl silan-
various types. riety of methods. ated silica.
Mode of Action of Antifoams 71

Particle Any other


size additives Application Date Ref.

0.007- Alkaline aqueous solns, 1971 94


0.025 particularly aqueous
micron. paints, latex systems,
laundry and detergent
products.

I . Whole mixed with Fabric washing 1974 95


sodium tripolyphosphate formulations.
to produce free-flowing
powder. This is the par-
ticular provision of the
patent. 2. Emulsifying
agents, maid inhibitors,
etc.

Not >0.l Antifoam incorporated Detergent formulations 1975 96


micron, in detergent impermea- for automatic clothes
preferably ble water soluble or dis- and dishwashing
0.01-0.02 persible carrier material machines.
micron. such as gelatin, polyeth-
ylene glycol, etc. This
material is in turn
coated with a water-sol-
uble granular material.
72 Garrett

TABLE 4 Some Examples of Antifoam Patents Concerning Mixtures of Hydro-


phobed Mineral Particles and Organic Liquids

Preferred
Actual concentration
Oils claimed to Particles claimed examples of particles/
be effective to be effective given % by weight

Kerosene, na- Aerogel, fume, or Silicas hydro- 3-20


phthenic mineral precipitated silicas phobed with sili-
oil, paraffinic min- hydrophobed by cone oil or with al-
eral oil, chlorinated any suitable kyl chlorosilanes +
naphthenic mineral method (but only most of the oils
oil, chlorinated par- silicone oil or alkyl claimed.
affinic mineral oil, chlorosilanes
liquid tri-fluorovi- mentioned).
nyl chloride poly-
mer, fluorinated
hydrocarbons.
Aliphatic or aro- Precipitated silica Precipitated silica 3-30
matic hydrocarbons having pH from 8- hydrophobed with
with at least six 10. Rendered hy- polymethylsiloxane
carbon atoms. drophobic by any and mixed with
suitable method but various mineral
only polysiloxane oils.
or alkyl (aryl or al-
icyclic) silanes
claimed. Type of
silica employed is a
special provision of
the patent.
Water-insoluble po- Silica subjected to 1. Polypropylene 1-10
lyalkylene glycol. heat and shear glycol + silica.
treatment in oil. 2. Polybutylene
Silica may also be glycol + silica
hydrophobed before 3. Polypropylene
mixing with oil by glycol + silica hy-
treatment with sil- drophobed with tri-
ane or alcohol. methyl silane.
4. Polypropylene
glycol + silica hy-
drophobed with
isobutanol.
Mode of Action of Antlfoams 73

Particle Any other


size additives Application Date Ref.

Most preferred Spreading agent claimed Probably mainly in- 1963 97


from 0.02-1 to be essential (anionic, tended for paper pulp
micron. cationic or nonionic mills although patent
surfactant). not restrictive in this
respect.

0.005-0.050 Need for a spreading Especially adapted for 1965 98


micron. · agent is eliminated-a paper pulp mill applica-
special feature of this tion although patent ap-
patent. Use of a spread- parently not restrictive
ing agent is claimed to in this respect.
reduce rather than in-
crease the effectiveness
of the antifoam.

Finely divided No specific application 1967 99


high surface quoted.
area (>50 m2
gm-1).
74 Garrett

TABLE 4 continued

Preferred
concentration
Oils claimed to Particles claimed to Actual of particles/
be effective be effective examples given % by weight

Aliphatic hydrocar- Any type of silica Mineral seal oil + Preferred 1-


bon or paraffin oil possessing reactive fume silica + di- 40 (but refer-
including mineral surface hydroxyl methyl ence made to
seal oil, kerosene, groups rendered dichlorosilane. the use of the
various light ali- hydrophobic with particles by
phatic fuel oils, gas dialkyl dihalosi- themselves
oils, paraffin Jane. Reaction i.e., 100).
waxes, etc. made in the oil.
Special claim of
patent concerns re-
action conditions
giving rise to di-
alkyl-substituted
cyclic siloxanes on
the surface of
silica.
Paraffinic and nap- Any type of silica Mineral seal oil 7-45
thenic mineral oils, possessing reactive with silica treated
cutting oils, kero- surface hydroxyl with various alkox-
sene, similar petro- groups rendered ysilicon chlorides.
leum fractions in- hydrophobic by re-
eluding food-grade acting with alkoxy-
mineral oils and silicon chloride
halogenated hydro- SiClm(OR)n Reac-
carbons. Synthetic tion made in the
oils also claimed to oil.
be suitable: ali-
phatic diesters, sili-
cate esters and po-
lyalkylene glycol or
their derivatives
(detailed descrip-
tion of viscosity,
density, etc., of
oils also specified).
Halocarbon or hy- Synthetic alkali Sodium or calcium 5-30
drocarbon fluid. metal or alkaline silicoaluminates hy-
earth metal silicoal- drophobed with
uminate rendered methyl chlorosi-
hydrophobic by re- lanes in paraffinic
action with halosi- mineral oil, chlo-
lane. Reaction ronaphthalene or
made in the oil. kerosene.
Mode of Action of Antifoams 75

Particle Any other


size additives Application Date Ref.
0.15-0.005 Emulsifiers and disper- Wide variety of indus- 1968 100
micron (ulti- sants may be added. trial applications
mate particle quoted. A specific
size but states claim concerns paper
that greatly pulp stock.
preferred are
nonaggregated
particles of
silica).

Less than 0.1 1. Need for spreading Paper pulp mill 1969 101
micron but agent is eliminated-a application.
preferably less special feature of this
than 0.05 mi- patent. Surface-active
cron (ultimate agent claimed to stabi-
particle size). lize the foam.
2. Alkylene oxide may
be added to react with
HCI formed when alk-
yoxy silicon chloride
reacts with silica (in the
oil).

<0.2 micron. Lewis base may be Paper pulp mills repre- 1970 102
This is a par- added to neutralize hal- sent a preferred
ticular provi- oacid formed in reaction application.
sion of the of halosilane with
patent. silicoaluminate.
76 Garrett

TABLE 4 continued

Preferred
Actual concentration
Oils claimed to Particles claimed to examples of particles/
be effective be effective given % by weight

Water-insoluble or- Aluminium oxide Colloidal alumin- Most pre-


ganic liquid se- reacted in situ with ium oxide prepared ferred from
lected from vegeta- alkali or alkaline by hydrolysis of al- 6-16.
ble oils, aliphatic earth hydroxide and uminium chloride
hydrocarbons, ali- fatty acid (with 6 in a flame + paraf-
cyclic hydrocar- to 24 carbon finic hydrocarbon
bons, halogenated atoms). oil + Ca(OHh +
aromatic hydrocar- variety of fatty
bons, long-chain acids.
alcohols, long-
chain esters and
amines (organic
liquid has boiling
point >65°C and is
liquid at room tem-
perature). A depen-
dent claim states
that liquid is
hydrophobic.
Water-insoluble or- Patent claim exclu- Silica hydrophobed 3-30
ganic liquid se- sively concerns the with dimethylpoly-
lected from kero- preparation of an siloxane and dis-
sene, naphthenic alkalized microfine persed in na-
mineral oil, paraf- precipitated silica phthenic, mineral
finic mineral oil, (which may be hy- or mineral seal
chlorinated nap- drophobed by any oils.
thenic mineral oil, method).
chlorinated paraf-
finic mineral oil
and liquid difluo-
rovinyl chloride
polymer. (Detailed
description of vis-
cosity, volatility,
etc., also
specified.)
Mode of Action of Antifoams 77

Particle size Any other additives Application Date Ref.


Alumina <15 (i) Water may be added Probably intended for 1972 103
micron pre- (forming water in oil paper pulp mills al-
ferred 0.01- emulsion). though reference not
1.3 micron. (ii) Surfactant may be made to such an appli-
added. cation in the claims.

<0.05 micron Catalyst to promote re- Particularly paper pulp 1973 104
(ultimate parti- action of silicone with mill application.
cle size). silica (e.g., tin octoate).
Addition of spreading
agents or other surfac-
tants usually not helpful
except occasionally
when silicone oil load-
ing of silica low.
78 Garrett

TABLE 5 Some Examples of Antifoam Patents Concerning Mixtures of Intrinsi-


cally Hydrophobic Particles and Organic Liquids

Actual Preferred concen-


Oils claimed to Particles claimed examples tration of parti-
be effective to be effective given cles/% by weight

Water-immiscible Polyamide of N,N' -distearyl 0.05-10


organic liquid polymethylane ethylene diamide
boiling above polyamine having + white spirit.
l00°C, e.g., a 2-12 methylene
hydrocarbon. groups and a car-
boxylic acid of
the group consist-
ing of aliphatic
and cycloaliphatic
monocarboxylic
acids, each acyl
group containing
11-18 carbon
atoms.
Organic liquid is Finely divided 1. Polypropylene 2-25
nonsolvent for poly-a-olefin + xylene
particle and is im- polymers such as 2. Polypropylene
miscible with polypropylene, + commercial ar-
water. Polar + polyisobutyl- omatic hydrocar-
nonpolar are ene,etc. Also bon mixture
listed. That is sat- claims thermo- 3. Polypropylene
urated unsaturated plastic polyesters + mineral oil +
aliphatic hydro- such as nylon. surfactant
carbons. Polar 4. Polyethylene
materials include tetraphthalate +
alcohols, esters, mineral oil, etc.
ketones, chlori-
nated aromatic
hydrocarbons,
fluorinated
hydrocarbons.

This terminological distinction between polydimethylsiloxane hydrocar-


bon-based systems suggests a complete mechanistic dissimilarity which seems
unlikely. It is more probable that the terminology simply reflects the relative
costs of the oil and particulate ingredients!
It is clear from the patent literature that an antifoam mechanism invoking
Mode of Action of Antifoams 79

Particle Any other


size additives Application Date Ref.

"Fine particles" Anionic detergent 1955 105


(prepared by heat- solutions. (has no
ing till dissolved detrimental effect on
and rapidly cool- the wetting or deter-
ing or by gent power of the
milling). solution or on its
content of active
material).

0.02-50.0 micron Surface active agent Reference made to 1972 106


but preferred size may be added in or- paper pulp mill ap-
0.2-5.0 microns. der to provide "an plication only.
improvement in the
property of these dis-
persions to spread at
the air-water
surface."

Marangoni spreading of the oil (sec Fig. 7) underlies some of the thinking
behind the development of these oil-based antifoams. It is, however, equally
clear that it represents an inadequate explanation for all reported phenomena.
Thus, Boylan [97], in a patent concerning the addition of hydrophobed silica
to organic liquids, claimed that the addition of a spreading agent (a soap for
TABLE 5 continued

Preferred
Actual concentration of
Oils claimed to Particles claimed examples particles/% by
be effective to be effective given weight
Hydrophobic or- Hydrates of fatty Mineral oil + 2-20
ganic solvents acid mixed salts i. Aluminium-
e.g., aliphatic cy- of polyvalent magnesium
cloaliphatic, hy- metals and/or of stearate.
droaromatic, aro- lower dibasic ii. Zinc-ethylene
matic amines. diamine stearate.
hydrocarbons, iii. Magnesium-
natural fatty oils zinc stearate
(e.g., olive oil),
silicone oils,
phosphoric acid
esters.
Mineral oil of Alipathic diamide N,N' -distearyle- 4-12
specified viscosity derivatives of po- thylene diamine
lymethylene + mineral oil.
diamine.
Hydrocarbon oil Amide which is Stearic diamide of 1-20
the reaction prod- ethylene diamine
uct of a polyam- + paraffin oil.
ine containing at
least one C2-C6
alkylene group
and a C 6 -C1s
fatty acid.
Mineral oil or es- Mono- or diester l . Mineral oil + 5-15
ters of unsatu- of hydroxystearyl hydroxy-stearyl (Particles may
rated fatty acids alcohol with satu- monobehenate + however dispersed
with mono or po- rated fatty acid or various other ad- in aqueous phase
lyhydric alcohols, hydroxy-fatty ditives, etc. directly and dem-
liquid fatty acids acid. onstrate antifoam
or alcohols, ter- properties by
pene themselves).
hydrocarbons.
Mineral oils, fatty Polyethylene hav- As claimed. 0.5-15
oils, fatty acids, ing molecular
tetraisobutylene. weight from 500
to 25,000.
Particle
size Any other additives Application Date Ref.

Nonionic emulsifier General industrial 1972 107


application.

15-20 microns 1. Spreading agent Paper pulp mill 1973 108


(by Hegman grind 2. Silicone oil application
gauge) (dependent claims)

Prepared by i. Oil-soluble General industrial 1973 109


heating till organic polymer application
particles dissolve ii. a fat
in oil followed by iii. silicone oil
quick chilling and
homogenizing to
"obtain smaller
particles. "
"Finely divided" Silicone oils, Paper pulp mills, 1975 110
nonionic ethoxylated manufacture of
surfactants, metal dispersions of
stearates, etc. plastics, etc.

Finely divided Emulsifier. General application 1975 111


"probably in the for aqueous systems.
range of 0.1
micron." Prepared
by dissolving in
oil at high
temperature and
then cooling to
obtain optimum
particle size.
82 Garrett

TABLE 6 Initial Values of Entry and Spreading Coefficients for Polydimethyl


Siloxanes

Siloxane Average Tempera-


(ex DOW Viscosity mol. ture 'YAO 'Yow
Corning) (mPa s) weight" (QC) (mN m- 1) (mN m- 1)

MS200 l.O 2.3 x 102 24 16.4 42.7


MS 200 5.0 6.7 x 102 24 19.3 42.8
DC 200 10.0 20 20. l 39.5
DC 200 50 20 20.8 34.3
DC 200 1,000 2.5 x 104 20 21.2 36.9
DC 200 60,000 7.0 x 104 20
DC 200 100,000 8.5 x 104 20 21.3
'Number average
bDircct measurement of initial spreading pressure using Wilhelmy plate.

example) to the oil is necessary in order to allow the water-insoluble organic


liquid to spread at the air-water surface. In later patents concerning the same
system, however, it is claimed that the addition of a spreading agent is un-
necessary. Buckman [101], for example, states that Boylan's product is not
entirely satisfactory because the presence of a surface-active agent in the
system tends to stabilize the foam. In another patent also concerning this
system, Miller [104] states that the addition of spreading agents may give
an improvement only if the surface treatment of the silica is inadequate.
Boylan apparently realized the error of his ways and eventually issued a
patent [112] in which the spreading agent was claimed as an optional additive.

B. Spreading Behavior of Antifoam Oils on Water


As we have seen there appears to be a consensus that necessary properties
of antifoam oils include insolubility and a tendency to emerge into the air-
water surfaces of foam films as marked by E > 0 (see Sec. III.B.). How-
ever, we have also seen that the spreading behavior of the oil may play a
role in the mode of action of the antifoam. Before examining this possibility,
we consider the relevant properties of typical oils used in oil-particle anti-
foams. In the main this means polydimethylsiloxanes and hydrocarbons.
Polydimethylsiloxanes are characterised by low air-oil surface tensions of
~ 20 mN m- 1 at ambient temperatures. Values of 'YAo and the oil-water
surface tension 'Yow for typical oils are shown in Table 6. Entry and spread-
ing coefficients are also presented. These are calculated by assuming a value
of 'YAw equal to that of distilled water. They therefore represent nonequilib-
rium "initial" values of E and S. Equilibrium values of E" and S" should
Mode of Action of Antifoams 83

Water 8 x 10- 3 M SDS Solution ('YAw = 39.0 mN m- 1)


E s 'Yow E s
(mN m- 1) (mN m- 1) (mN m I) (mN m- 1) (mN m- 1) References

+98.2 +12.8 9.8 +32.4 +12.8 113


+95.4 + 9.8 12.0 +31.7 + 7.7 113
+90.9 + 11.9 48
+85.0 +16.4 48
+87.2 +13.4 + 7.3h 48
+10.7b
+ l l.Ob

be calculated with the values of 'YAw obtained after contamination with po-
lydimethylsiloxane (which forms a monolayer at the air-water surface). Un-
der those circumstances it is possible that S" < 0 and the oil will not spread
but will form a lens (where we still have E" > 0) [32,33).
If a small droplet of polydimethylsiloxane oil is inserted into the air-water
surface of distilled water in a narrow through, it spreads as a duplex film
at a rate given by Eq. 19. This equation implies that the rate is dependent
upon S but is independent of the viscosity of the polydimethylsiloxane. In
contrast to viscosity, Sis only weakly dependent upon molecular weight for
high molecular weights (see Table 6). Therefore the spreading rate is es-
sentially independent of molecular weight provided Eq. 19 is valid. We have,
however, observed that for number average molecular weights > 2.5 X 104
(and viscosities> 1000 mPa s) Eq. 19 tends to overestimate spreading rates
which become dependent upon polydimethylsiloxane viscosities at high mo-
lecular weights [114]. After spreading, the film appears to very slowly (de-
pending on the viscosity of the oil) break up into oil lenses. A second droplet
of oil introduced to any free air-water surface formed as the film breaks up
does not spread but forms a lens.
The effect of adding polydimethylsiloxane oils to the surfaces of surfac-
tant solutions is somewhat less well documented. Again, however, the oil
usually spreads as a duplex film and subsequently slowly breaks up to form
lenses. A second droplet added to the siloxane-contaminated surface does
not spread but forms a lens, so that equilibrium values of the spreading
coefficient S" < 0. This is illustrated in Fig. 28 for a 5 x 10- 3 M solution
of 6-phenyldodecylbenzenesulfonate in 4 x 10- 2 M NaCl to which two drops
of 1000 mPa s polydimethylsiloxane were added in sequence.
Estimates of the initial values of E and S for polydimethylsiloxanes on
sodium lauryl sulfate solutions calculated from the surface tension results of
84 Garrett

'YAw I mNm-1

~first drop polydimethylsiloxane added

26

-
drop spreads to
form duplex film

25

24

second drop polydimethylsiloxane


duplex film thins added forms lens
and lenses form

23

0 60 120 180 240 300 '360 420 480 540 600


Time Is

FIG. 28 Spreading behavior of polydimethylsiloxane on 5 x 10- 3M sodium 6-


phenyldodecylbenzenesulfonate in 4 x 10- 2M NaCl at 25°C (polydimethylsiloxane
of 1000 mPa s (DC200 ex Dow Coming); drops of -0.01 cm3 added to dish of lO
cm diameter). (From Ref. 114.)

Kannelopoulos and Owen [113] are given in Table 6. Spreading rates may
well be lower than values predicted by Eq. 19 because 'Yow increases as the
oil-water interface expands and because compression of the surfactant mono-
layer ahead of the spreading front may also occur. There is, however, a
paucity of data concerning the spreading behavior of polydimethylsiloxane
oils on surfactant solutions. A detailed study of the spreading behavior of
polydimethylsiloxane on surfactant solution would serve to educate the de-
bate concerning the relevance of that property for antifoam behavior.
The relevant surface properties of hydrocarbons are shown in Table 7. It
is seen that homologues higher than n-heptane exhibit initial spreading coef-
TABLE 7 Initial Values of Entry and Spreading Coefficients for Hydrocarbons on Water

Temperature
Hydrocarbon (QC) 'YAo(mN m-l) 'Yow(mN m- 1) E"(mN m- 1) s•(mN m- 1) Ref.

n-pentane 20 16.0 50.2 +107.2 +6.8 115


n-hexane 20 18.4 50.8 +105.4 +3.8 115
n-heptane 20 20.l 51.2 + 104.1 +1.7 115
ClO
U1
n-octane 20 21.7 51.7 + 103.0 -0.4 116
n-decane 20 23.8 52.3 +101.5 -3.l 116
n-dodecane 20 25.4 52.8 + 100.4 -5.2 116
n-tetradecane 20 26.5 53.8 +99.8 -6.8 116
n-hexadecane 20 27.4 53.8 +99.4 -8.2 116
Typical mineral oil 25 29.7 44.9b +87.5 -2.3 40
•Taking 'YAw = 73.0 mN m-• at 20°C and 72.3 mN m-• at 25°C.
•Low, presumably due to polar impurities.
86 Garrett

ficients S < 0 on distilled water at near ambient temperatures. For these


compounds E > 0 and lens formation occurs. Hydrocarbons used for anti-
foams tend to be mixtures of high molecular weight such as mineral oils (or
"white" oils, etc.). These, too, do not spread on distilled water at ambient
temperatures.
However, spreading on distilled water of all hydrocarbons including min-
eral oils can be induced by dissolution of oil-soluble surfactants in the oil
to reduce 'Yow· Such surfactants are often present as impurities in hydro-
carbon oils.
Hydrocarbons which do not spread on distilled water can spread on sur-
factant solution surfaces even when impurities are absent from the hydro-
carbons. Thus, certain hydrocarbons will exhibit a positive initial spreading
coefficient on surfactant solutions. Such hydrocarbons first spread as a du-
plex film, which subsequently breaks up to form lenses in equilibrium with
surfactant monolayer contaminated with hydrocarbon. However, the results
of A veyard et al. [ 117, 118) suggest that if the hydrocarbon has a sufficiently
long chain length relative to that of the surfactant, then this phenomenon
does not occur and the initial spreading coefficient becomes <0. We have
found, for example, that the mineral oil used to obtain the foamability results
shown in Fig. 19 does not detectably (using a Wilhelmy plate) reduce the
surface tension of the relevant sodium (C 10-C 14) alkylbenzene sulfonate
solutions (40).
The effect of surfactant in the aqueous phase on the initial dewetting and
spreading behavior of hydrocarbons can be predicted if certain assumptions
are valid. Thus, suppose that the surfactant is ionic with a single charge,
does not dissolve in the oil, and is not adsorbed at the oil-air surface (rea-
sonable assumptions for anionic and cationic surfactants in the absence of
added electrolyte). Then if we use the Gibbs equation, we may write for
the entry coefficient
dE d("'!AW +'Yow - 'YoA) = 2RT(- f - f ) (48)
dln a± d In a± Aw ow

and for the spreading coefficient

dS d("'/Aw - 'Yow - 'YoA) = lRT(- f + f ) (49)


dlna± dlna± AW OW

Where f AW and f OW are the SUrface excesses at the air-Water and OiJ-water
surfaces of the surfactant solution, and a± is the mean ionic surfactant ac-
tivity. Contact angle measurements have shown that adsorption at the hy-
drocarbon solid-water surface is equal to adsorption at the air-water surface
for anionic surfactants [119). This is a consequence of the dominance of the
Mode of Action of Antifoams 87

hydrophobic effect in determining adsorption behavior. It seems reasonable


to expect that adsorption at the air-water and the liquid hydrocarbon-water
surfaces will show the same behavior provided that the chain length of the
hydrocarbon is long so that it does not contaminate the air-water surface.
Therefore we may substitute r AW = r OW into Eqs. 48 and (49) to find
dE
- - - = -4RTf AW (50)
dln a±
and
dS
---=0 (51)
dln a±
Addition of surfactant to the aqueous phase may reduce the entry coefficient
(and may even "wet-in" the hydrocarbon so that E < 0). It will, however,
only induce spreading in a hydrocarbon which does not spread on distilled
Water if the assumption f AW = f OW is Violated (which is presumably the
case for shorter-chain-length hydrocarbons, for which we must have row >
f AW before hydrocarbon Contamination Of the Surfactant monolaycr, SO that
dS/d ln a± > 0 from Eq. 49). We have observed that the spreading coef-
ficient for a mineral oil on solutions of anionic surfactant is essentially un-
affected by surfactant concentration and is similar to that calculated for dis-
tilled water where S < 0, so that l\w = fow and Eq. 51 is satisfied [40].

C. Role of the Oil in Synergistic Oil/Particle


Antifoams
Mixtures of hydrophobed silica and polydimethylsiloxane are effective an-
tifoams for aqueous solutions of anionic surfactants. However, the oil alone
is not. Such oils can, as we have seen, spread at both the air-water surface
and air-surfactant solution surfaces.
Ross and Nishioka [84] have examined the stability of air bubbles (of
~0.8 cm radius) released under monolayers of polydimethylsiloxane spread
on both distilled water and surfactant solution. With a distilled water sub-
strate bubbles were more stable than in the absence of polydimethylsiloxane
(presumably as a result of reduced film draining rates due to lack of mobility
of the polydimethylsiloxane-contaminated surface). Similar findings are re-
ported by Trapeznikov and Chasovnikova [120]. With a surfactant solution
substrate the presence of a spread polydimethylsiloxane monolayer did not
diminish bubble stability. All of this suggests that displacement or contam-
ination of a surfactant monolayer with polymethylsiloxane to produce un-
stable foam films cannot be the role of this material.
88 Garrett

Anti foam
efficiency Initial Spreading
Coefficient I mNm-1
200 20

10
100

0.001 0.01 0.1 1.0


concentration SOS (%wt/wt)

FIG. 29 Comparison of initial spreading coefficient S and antifoam efficiency for


polydimethylsiloxane/hydrophobed silica antifoam on sodium dodecylsulfate solu-
tion. (From Ref. 30.)

Another possibility must be foam film rupture induced by spreading (see


Sec. III.C.). However, Kulkarni et al. [81] describe an experiment where
a drop of polydimethylsiloxane is added to foam formed from sodium lauryl
sulfate in a tray. This caused no foam destruction despite the obvious in-
ference that spreading from the droplet would occur, which would in turn
cause collapse if the spreading mechanism is effective in this context. More-
over, although Kulkarni et al. [30, 81] claim that antifoam action usually
requires a positive value of the initial spreading coefficient, they show that
antifoam efficiency does not correlate with the magnitude of S [30]. A com-
parison of antifoam efficiency with S for mixtures of hydrophobed silica/
polydimethylsiloxane is presented in Fig. 29. Here the increase in initial
spreading coefficient at high sodium lauryl sulfate concentration (where pre-
sumably S increases due to solubilization of dodecanol impurity in micelles)
contrasts with declining antifoam efficiency. However, as suggested by Ew-
ers and Sutherland [26], spreading would be expected to become more sig-
nificant for a greater spreading coefficient. Thus, the shear force applied to
the intralamellar liquid is dS/dx., where x is the distance of the spreading
layer from its source in the plane of the air-water surface. Higher initial
Mode of Action of Antifoams 89

values of S should therefore mean higher shear forces and a greater prob-
ability of foam film rupture.
It is possible, however, to query the relevance of the initial spreading
coefficients used by Kulkarni et al. [30] in Fig. 29. The method of foam
generation used release of gas through a glass frit. Any foam volume col-
lapse due to the effect of antifoam must involve the continuous air-water
surface which exists at the top of the resulting foam column. It seems rea-
sonable to suppose that accumulation of polydimethylsiloxane will occur on
that surface even before the experiment begins. Moreover, it seems probable
that silicone contamination of bubbles may have occurred by the time they
have reached the upper surface of the foam. On the whole, then, use of
initial spreading coefficients rather than equilibrium spreading coefficients
would appear to have a dubious basis. Equilibrium spreading coefficients
will be :50.
As we have seen, hydrocarbons also exhibit antifoam behavior when mixed
with hydrophobic particles. Here, however (provided precautions are taken
to eliminate contamination by oil-soluble surfactant), it is possible to achieve
antifoam effects for oil-particle mixtures where the initial spreading coef-
ficient for the oil is unequivocally <0. Thus, a detailed study of the spread-
ing behavior of the mineral-oil-based antifoam, used to obtain the foam re-
sults shown in Fig. 27, on solutions of sodium (C 10 - C 14) alkylbenzene
sulfonate revealed no evidence of spreading under either initial or dynamic
conditions (where the air-water surface is expanded in a controlled manner)
[40].
We therefore arrive at the conclusion that the role of the oil, if it is hy-
drocarbon, need not concern spreading at the air-water surface. If it is po-
lydimethylsiloxane, then, although the oil may spread, the evidence for a
role for that spreading is at best equivocal and does not support a mechanism
of film rupture by surface tension gradient induced shear. It is tempting then
to infer that spreading of the antifoam oil, irrespective of type, is not a
necessary condition for antifoam mechanism with mixtures of oils and par-
ticles. In this we differ from Kulkarni et al. [30,81], who state that a positive
initial spreading coefficient is a necessary, but not a sufficient, property of
such oils.
As we have seen, oils for which E > 0 and which do not spread at the
air-water surface form lenses. If the oil lenses are small, then gravitational
forces may be neglected and the configuration adopted is that shown in Fig.
30a. Here the air-water surface is everywhere planar, and both the oil-water
and oil-air surfaces form spherical segments. The angles formed by the three
fluid-fluid surfaces at the three-phase contact line are determined by Neu-
mann's triangle of forces as shown in the figure.
90 Garrett

'YoA

air 'YAW
water I

oil 8* /I

--
I
I. -
I
I
(a)
,I
I

'Yow

6.PoA
air

water oil

air

(b)
FIG. 30 Oil lens and definition of 0*: (a) oil lens at air-water surface illustrating
Neumann's triangle of surface forces; (b) oil lens bridging aqueous foam film. (From
Ref. 12 l.)

Garrett [121] has shown that if such an oil lens bridges a plane-parallel
foam film by emerging into both air-water surfaces then no configuration
of mechanical stability is available to the lens if the angle 0*, formed by
the tangents to the air-water and oil-water surfaces, is >90°. The condition
0* > 90° is satisfied provided we have
)'~w + )'~w - )'~A > 0 (52)
Condition 52 gives rise to unstable bridging configurations even when the
air-water surface is everywhere planar as shown in Fig. 30b. Here it can be
shown that the Laplace pressure drop across the oil-air surface /:,.PoA cannot
equal that across the oil-water surface Mow· In this situation /:,.P 0 A > 1~.P 0 w,
and an unbalanced capillary force will cause enhanced rates of drainage in
the aqueous film away from the lens.
Mode of Action of Antifoams 91

curvature of surface
implies capillary force
driving water away
from lens

two three-phase
contact lines
coincide so oil
lens dewetted to
leave hole

FIG. 31 Foam film collapse induced by bridging oil lens. (From Ref. 39.)

If, on the other hand, the air-water surface is nonplanar, then a config-
uration similar to that shown in Fig. 31 occurs when condition 52 is satis-
fied. Here again there is a capillary pressure enhancing the rate of drainage
away from the droplet. In this case, however, it is possible, as suggested
by Frye and Berg [39], to envisage a process where the oil lens is completely
dewetted from the foam film to form a hole. The process is shown sche-
matically in Fig. 31. Here thinning of the aqueous film causes the lens to
expand and ultimately results in coincidence of the upper and lower three-
phase contact lines. It is immediately obvious that this process is analogous
to the dewetting of a cylindrical particle in a foam film and therefore requires
0* > 90°.
The process depicted in Fig. 31 would, however, be only possible for
comparatively low values of El*. As 0* ~ 180°, then the curvature of the
oil-air surfaces will of necessity become concave if the oil-water surface is
92 Garrett

to remain cylindrical. In this case the point of rupture could be at the center
of the expanding oil lens as the two concave air-oil surfaces approach one
another.
Condition 52 for formation of destabilizing bridging configurations by oil
lenses is strongly dependent upon 'YAw and 'YoA because, in general, 'Yow <.;
'YoA and 'Yow <.; 'YAw· Indeed, essentially all that is required for such behavior
is that the oil-air surface tension be less than the air-water surface tension
of the foam film surfaces. Such a condition is readily satisfied for polydi-
methylsiloxane oils but will not always be satisfied for hydrocarbon oils.
Some evidence that the behavior of hydrocarbon oil-based antifoams fol-
lows that predicted by condition 52 has been presented by Garrett et al.
[122]. This study concerned the effect of a mineral oil/hydrophobed silica
antifoam on the foarnability and foam stability of a homologous series of
sodium alkylbenzene sulfonate isomer blends. Of particular interest is the
effect of that antifoam on the stability of the foam several minutes after foam
generation ceased. The ambiguities associated with dynamic effects are then
minimized, and the air-water surface tension of foam films is reasonably
supposed to approximate the equilibrium surface tension of a free air-water
surface. In Table 8 we present the entry coefficient E, the spreading coef-
ficient S, and the bridging coefficient B, where
(53)
for each of the chain lengths of alkylbenzene sulfonates calculated by using
the equilibrium values of 'YAw and 'Yow· Here it is clear that everywhere E
> 0 and S < 0. However, for chain lengths > C 12 the bridging coefficient
B > 0 and therefore 0* > 90°.
In the absence of antifoam the foam volume of each of these solutions
of alkylbenzene sulfonate was stable for up to 1800 s after foam generation
ceased (although bubble disproportionation and drainage significantly af-
fected the appearance of the foam). A measure of the effectiveness of the
antifoam after foam generation has ceased is the ratio F(t = 960 s)/F(t =
0), where t is the foam age and where we remember F = volume of air in
foam in presence of antifoam/volume of air in foam in the absence of an-
tifoam. In effect, the ratio F(t = 960 s)/F(t = 0) is the ratio of the volume
of air in the foam in the presence of antifoam after 960 s to the volume of
air in the foam in the presence of antifoam immediately after foam gener-
ation has ceased. This follows because the foam volume is stable in the
absence of antifoam. The ratio F(t = 960)/F(t = 0) has the value of unity
if F does not change with the age of the foam because the antifoam does
not function under the near-equilibrium conditions then prevailing. It is com-
pared with Sand B in Table 8. Here we see that F(t = 960)/F(t = 0) <.; 1
for chain lengths <C 12 where B > 0. Therefore, significant foam film col-
TABLE 8 Comparison of Entry E, Spreading S, and Bridging B Coefficients with Mineral-Oil-Based Antifoama Effectiveness after
Foam Generationb Has Ceased for Sodium Alkyl Benzene Sulfonatesc
Antifoam
effectiveness
Chain F(t = 960 s)/
Length 'YAW (mN m- 1) 'Yow (mN m- 1) E (mN m- 1) S (mN m- 1) B (mN m- 1)2 F(t = 0 s)

33.0 6.5 +8.5 -4.5 +170.3 0.50


co ~
34.0 6.8 +9.8 -3.8 +241.2 0.33
w C10
31.7 6.75 +7.5 -6.1 +89.4 0.62
Cu
30.2 4.9 +4.1 -5.7 -24.9 0.94
C12
C13 28.8 4.0 +1.8 -6.2 -115.6 1.00
27.7 3.3 0.0 -6.6 -182.8 0.90
C14
"Antifoam is a mixture of mineral oil 'YAo = 31 mN m- 1 and hydrophobed silica.
&Foam measured using Ross Miles apparatus.
•A mixture of all possible isomers except the I-phenyl isomer.
Source: Ref. 122.
94 Garrett

lapse due to the antifoam after foam generation has ceased only occurs for
solutions where B > 0, and therefore 0* > 90°. This could represent some
evidence that foam film collapse can be induced by a bridging oil lens which
does not spread at the air-water surface. It does, however, ignore the role
of the particle.
The bridging oil lens then provides an explanation for the role of the
antifoam oil, which does not require that the oil spread at the air-water sur-
face. It fails, however, to address the central question of the role of the
particle.

D. Hypotheses Concerning the Role of the Particles


in Synergistic:: Oil/Particle Antifoams
In the past decade or so a number of different hypotheses have been ad-
vanced in explanation for the role of particles in synergistic oil-particle an-
tifoams. Much of the relevant work has concerned mixtures of oil with silica
where the requirement that the silica be surface treated to render it hydro-
phobic has been amply demonstrated.
Some of these hypotheses are essentially naive and are easily disposed
of. Thus, Sinka and Lichtman [123] attribute the role of the particle to in-
hibition of solubilization of the oil. However, it is easily shown that the
presence of concentrations of antifoam oil well in excess of the (usually
extremely low) amounts which are solubilized may still produce negligible
antifoam effect with aqueous solutions of, say, anionic surfactants [40 J. Povich
[83] shows that the increase in bulk shear viscosity of polydimcthylsiloxane
due to the presence of the silica is not responsible for the sevenfold increase
in antifoam effectiveness accompanying presence of that material in the oil.
Moreover, Ross and Nishioka [85] show that the presence of silica has no
effect on the surface shear viscosity of polydimethylsiloxanc. Garrett et al.
r401 have shown that although the presence of hydrophobed silica particles
can facilitate dispersal of the antifoam oil that is not the principal role of
the particles. Thus, mineral oil dispersions of essentially the same size dis-
tribution as hydrophobed silica/mineral oil dispersions reveal a markedly
different antifoam effectiveness. This is exemplified in Fig. 32, where the
relevant size distributions are compared with the respective antifoam effec-
tiveness as measured by F for a solution of a commercial sodium alkylben-
zene sulfonate.
The possibility that the spreading coefficient of polydimethylsiloxane oils
is modified by the addition of hydrophobic particles has been explored by
Povich [83]. Here the initial spreading coefficient for polydimethylsiloxane
has been shown to slightly decrease upon addition of hydrophobed silica
(despite the effectiveness of this silica in promoting the antifoam behavior
Mode of Action of Antifoams 95

Fraction of Particles of
given Size Range

0.5
(a) F =0.77 (b) F =0.12
(60 minutes (10 minutes
agitation) agitation)

0.4 0.4

0.3 0.3

0.2 0.2

0.1 0.1

0.5 10.0 20.0 0.5 10.0 20.0


5.0 15.0 25.0 5.0 15.0 25.0

droplet diameters I microns

FIG. 32 Comparison of antifoam effectiveness F with size distribution of (a) min-


eral oil, (b) mineral oil/hydrophobed silica (90/JO by weight). Here the size dis-
tributions were adjusted by altering the agitation time so that they arc approximately
equal. (From Ref. 40.)

of the oil). Povich [83] attributes this to adsorption of oil-soluble surface-


active impurities on the silica. Hydrophobed silica has also been shown to
be without significant effect on the spreading behavior of a mineral oil [40].
An often-quoted mechanism for mixtures of hydrophobic oils and parti-
cles is that due to Kulkarni et al. [30,81). These authors claim that the oil
spreads over the air-water surface exposing the particles to the aqueous so-
lution. Adsorption of surfactant onto the surface of the particles is then sup-
posed to occur, rendering the particle hydrophilic so that particles are pro-
gressively extracted from the oil into the aqueous phase. Rapid local depletion
96 Garrett

of surfactant in the aqueous film is then in turn supposed to produce a "sur-


face stress" which renders the foam film unstable so that rupture occurs.
There are a number of problems with this mechanism. The first problem
clearly concerns the requirement that the oil spread at the air-water surface.
We have already established that this is not a necessary property of antifoam
oils. A second problem is conceptual-depletion of surfactant in the foam
film by adsorption on particles will give rise to an increase in surface ten-
sion, which will in tum produce a Marangoni flow in the direction of the
foam film. This will tend to enhance film stability. Finally it is clear that
an aspect of the mechanism is an intrinsic tendency for the particle to be
removed from the oil phase to the aqueous phase. However, it has been
shown [40] that addition of mineral oil to a dispersion of hydrophobed silica
in sodium alkylbenzene sulfonate solution produces an enhancement of an-
tifoam effectiveness. The results are given in Table 9. This is clearly not
consistent with a process of removal of particles from oil having a central
role in the synergistic mode of action of the antifoam. On the whole, then,
we are forced to conclude that the mechanism proposed by Kulkarni et al.
[30,81] is probably wrong.
Dippenaar [52] has suggested that the particle may have a central role in
the mode of action of the antifoam. He supposes that the particle functions
by a dewetting mechanism similar to that outlined in Sec. IV.B. The oil is
supposed to adhere to the particle to yield a higher contact angle than would
otherwise prevail. Dippenaar [52] illustrates this mechanism by comparing
the contact angle and effectiveness at film rupture of sulfur particles before
and after contamination with liquid paraffin. This mechanism would, how-
ever, be most effective in the case of rough particles where the oil may
adhere in the rugosities to increase the contact angle at the air-water surface.
It is, however, difficult to see how this mechanism could function where
the oil forms the overwhelming proportion in the antifoam and where the
particle is preferentially wetted by the oil. In this case, entities where a
rough particle is embedded in an oil droplet with only minimal exposure of
the particles to the aqueous phase will be favored. This issue is addressed
in more detail below (Sec. V.E.).
Frye and Berg [39] have proposed that if the oil completely wets the
particles then the size of those oil droplets containing particles will be de-
termined by the size of the particles. Sufficiently large particles should there-
fore reduce the time required for foam films to thin to the point where bridg-
ing and foam film collapse can occur according to the mechanism suggested
by condition 52 and shown in Fig. 31. However, the presence of particles
which produce synergistic foam behavior does not necessarily mean larger
oil droplets, as we have shown in Fig. 32. Moreover, reduced thinning time
does not necessarily imply increased frequency of foam film collapse be-
TABLE 9 Effect of Adding Mineral oil to a Dispersion of Hydrophobic Silica in 0.5 g dm- 3 Sodium Alkyl Benzene Sulfonate• Solution
(0.03 g of hydrophobed silicab in 25 cm3 of surfactant solution. Foam generated by shaking 100 cm3 cylinder)

After 15 s After 15 s After 15 s


After 15 s After 2 shaking for shaking for After 1.5 h shaking for
shaking h standing second time 'third time standing fourth time
0.41 0.36 0.55 0.18 0.0 0.15
!S 0.43 0.40 0.52 0.10 0.0 0.11

Approximately 0.03 ± 0.0lg mineral oil added

'A commercial alkylbenzene sulfonate of C 10-c,. chain length (Dobs 055).


•A silanized Aerosil 200.
Source: Ref. 40.
98 Garrett

cause larger antifoam entities means fewer entities. Thus, we remember the
finding of Dippenaar [52] that the mass of antifoam required to remove a
given proportion of foam is proportional to the size of the antifoam-smaller
entities therefore mean a higher overall antifoam efficiency. The hypothesis
of Frye and Berg [39] also ignores the usual observations [40,123] that the
hydrophobic particles are not completely wetted by the oil but adopt a finite
contact angle Bow at the oil-water surface so that
90° <Bow< 180° (54)
where 80 w is measured through the aqueous phase.
On the whole, then, we are left with the conclusion that no published
hypothesis concerning the role of the particles in synergistic oil-particle an-
tifoams is entirely satisfactory.

E. Capture of Oil Droplets by Bubbles and the Role


of the Particles
1. Experimental Observation
As we have seen, the particulate component of an oil-particle antifoam has
a contact angle at the oil-water surface so that condition 54 is satisfied. This
is in fact the condition required for water-in-oil Pickering emulsion for-
mation where particles adhering to the oil-water surface ensure high stabil-
ity. Indeed a characteristic of oil-particle antifoams is that if equivolume
amounts of the antifoam and the solution to be defoamed are shaken together
a water-in-oil emulsion invariably forms [40,123]. If, however, the oil alone
is shaken with the same solution, an oil-in-water emulsion usually results.
Clearly, then, there is strong evidence to suggest that the particles in the
oil-particle antifoam adhere to the oil-water surface. Measurements of con-
tact angles on compressed disks of particles or on smooth plates of repre-
sentative materials seem to confirm this conclusion and imply that condition
54 is satisfied [40, 123 J. A finding by Frye and Berg [39] that the receding
contact angle Bow is 180° for solutions against hexadecane on hydrophobed
glass plates supposedly representative of the hydrophobed glass antifoam
ingredient used by these authors would appear to contradict these findings.
However, Frye and Berg f39] do not present emulsion evidence where the
presence of their antifoam particles should not invert the emulsion behavior
of the solution-hexadecane system (which should be oil-in-water) if their
contact angle measurements are correct.
The balance of evidence then suggests that the particulate component of
the antifoam adheres to the oil-water surface with a contact angle which
satisfies condition 54. A schematic illustration of such an entity is shown
in Fig. 33. It seems probable then that the location of the particles at the
Mode of Action of Antifoams 99

--
'
-
' 8ow

water

oil

FIG. 33 Composite antifoam entity comprising mixture of oil and spherical hy-
drophobic particles with 90° < 00 w < 180°.

oil-water interface is the key to their role. Kulkarni et al. [ 1241 have em-
phasised the importance of long-range electrostatic repulsion forces between
antifoam entities and bubbles in contributing to antifoam behavior with hy-
drophobed silica/polydimethylsiioxane antifoams. These authors present some
evidence of diminished antifoam effectiveness accompanying increasing zeta
potentials in sodium lauryl sulfate solution with increasing concentration. It
seems probable then that the role of particles at the oil-water surface will
concern the (often electrostatic) forces between bubbles and antifoam entities.
Further evidence concerning this proposition is given if we note that hy-
drophobic particles are often not added to polydimethylsiloxanes when these
materials are used as antifoams for nonaqueous systems. In a study of poly-
dimethylsiloxane anti foam behavior in lube oils, Shearer and Akers [ 17]
have shown that electrostatic interactions are essentially absent in that system.
Anfruns and Kitchener [75], as we have seen, have shown that the capture
of particles by bubbles is greatly facilitated by the presence of asperities.
This has been attributed to the penetration, by an asperity, of the energy
barrier associated with long-range (electrostatic) repulsion forces between
100 Garrett

D Mineral oil + hydrophobed silica 90/10 by weight

• Mineral oil
Fraction air/water/oil
films ruptured


1.0

0.5

10 20 30
Time I seconds

FIG. 34 Effect of hydrophobed silica on time of emergence of mineral oil droplets


into air-water surface of 0.5 g dm- 3 sodium (C 10-C 14) alkylbenzene sulfonate so-
lution (Dobs 055). Drop volumes 0.01 cm3, hydrophobed silica Dl7 ex Degussa.
(From Ref. 40.)

particles and bubbles (see Sec. IV.E.). It seems reasonable to attribute a


similar role to particles at the oil-water surface. Thus, in effect, the particle
should facilitate rupture of the aqueous film separating an oil droplet from
the air-water surface. This hypothesis has been tested by examining the ef-
fect of hydrophobed silica on the time of emergence of mineral oil droplets
into the air-water surface of the solution of a sodium (C 10-C 14) alkylben-
zene sulfonate [40] used to obtain the foam results shown in Fig. 27. The
droplets were of volume -0.01 cm3 • Results are presented in Fig. 34. Here
it is clear that the particles significantly reduce the emergence time for the
oil droplets. A similar result has been obtained for oil droplets coalescing
with a layer of oil on the surface of a solution [40]. This then implies that
the emulsion behavior found with these particle-oil mixtures is determined
by a tendency of the particles to rupture the aqueous film between oil drop-
Mode of Action of Antlfoams 101

lets. Essentially the same explanation for the effect of particles on emulsion
behavior has been offered by van Boekel and Walstra [125] and by Mizrahi
and Bamea [126).
It would seem that the role of the particles in oil-particle antifoams is to
facilitate the emergence of oil droplets into the air-water surface. Thus, cap-
ture of an oil droplet by a bubble will not occur if the time taken for the oil
droplet to be swept along streamlines around the bubble is less than the time
for emergence. The presence of particles will therefore increase the prob-
ability of capture. Here, however, we would expect the time scales to be
different under such dynamic conditions than those revealed by Fig. 34.

2. Spherical Particles
The process of emergence of an oil droplet into the air-water surface will
clearly be initiated by penetration of the energy barrier due to long-range
forces. However, ultimately the particle must facilitate hole formation in the
oil-water-air film in much the same manner as particles rupture air-water-
air films. The process is illustrated in Fig. 35 for a spherical particle sat-
isfying condition 54. Here we suppose that the particle is small so that the
effect of gravity on the shape of the oil-water surface may be neglected. It
is easily seen that for the particle to rupture the film we must have

(55)

If the particle also satisfies condition 54, then condition 55 implies that rup-
ture of an oil-water-air film by a spherical particle can occur if 0Aw < 90°.
Such a particle will not, as we have seen, rupture symmetrical air-water-air
films (for which we require eAW > 90°). We therefore find that spherical
particles which satisfy conditions 54 and 55 will promote the emergence of
an oil droplet into the air-water surface without exhibiting any antifoam be-
havior when used alone. If the oil satisfies condition 52, then we have a
clear explanation for the antifoam synergy which Frye and Berg [39] report
for mixtures of hydrophobed glass spheres and hexadecane. Thus, the par-
ticle alone will not function as an antifoam if the contact angle at the air-
water surface is <90° and the oil will not function because of low probability
of emergence into the air-water surface. Particles in the oil facilitate the
latter process, enabling the resultant oil lens to bridge an aqueous film and
participate in the foam film rupture mechanism shown in Fig. 31.
Inversion of the tendency of the oil to form oil-in-water emulsions by the
presence of spherical particles is also easily understood. Thus, if the par-
ticles are preferentially wetted by the oil so that they satisfy condition 54
then they may rupture aqueous films between oil droplets in much the same
102 Garrett

Film ruptures as three


phase contact lines become
coincident

I
air .-- - - 1
I
<II /Sow
/l/lll/l);/l/l
oil

FIG. 35 Rupture of air-water-oil film by spherical particle with 90° < 60 w < 180°
and OAw > 180° - 60 w so that OAw < 90°.

manner as spherical particles, with 0Aw > 90°, rupture aqueous films be-
tween air bubbles (Fig. 12 and 14).
3. Smooth Particles with Edges
In the case of smooth particles with edges then, as with air-water-air films,
a more complex set of conditions for rupture of oil-water-air films arises.
With an axially symmetrical particle defined by the angle eP and shown in
Fig. 18 we have only one configuration which unambiguously yields film
rupture if eAW < 90°. Thus, if we have
(56)

then a configuration similar to that of B in Fig. 18 will occur where the oil-
water surface replaces the air-water surface and the particle is half immersed
in oil. The configuration is depicted in Fig. 36. A particle adopting such a
Mode of Action of Antifoams 103

q5 Film rupture

t
/

t air

FIG. 36 Effect of axially symmetrical particles with edges on stability of oil-water-


air film with 0P < 00 w < 180° - 0P and 0Aw 0P.

configuration at the oil-water surface will rupture any oil-water-air film pro-
vided that 0Aw > 0P. This condition is of course the same as that (i.e.,
condition 29) required for the particle alone to cause air-water-air foam film
rupture. Oil-particle synergy would therefore not be apparent with particles
of this geometry. Such particles are not readily prepared. However, similar
conclusions are relevant for cubic crystalline particles where the same film
rupture configuration as at the air-water surface (see Sec. IV.B.) is possible
at the oil-water surface. An example of a cubic particle is a galena crystal
rendered hydrophobic by suitable surfactants. Assuming perfect crystal ge-
ometry, such particles should be equally effective at rupturing oil-water-air
films as for rupturing air-water-air foam films (provided 0Aw > 45°). Syn-
ergy would therefore be absent. However, the probability of achieving the
required diagonal orientation (see Fig. 16b) at the relevant interfaces will in
general be different, and a significant incidence of crystalline imperfections
104 Garrett

would mean additional edges and asperities. All of this could mean that
synergy may still be apparent.
As we have seen, more complex particles with straight edges can be mod-
eled by projection of the cross section shown in Fig. 18 in a direction normal
to the plane of the paper (see Fig. 19). In Table 10 we present the config-
urations such a particle could adopt at the oil-water interface if condition 54
is satisfied and if eP = 30°. The conditions required for 0Aw if the particle
is to cause rupture of oil-water-air films and facilitate emergence of the oil
droplet into the air-water surface are also deduced. Here we see that three
configurations give rise to unambiguous rupture of oil-water-air films with
0Aw < 90°. The remaining configurations will only affect rupture of such
films if the critical thickness of rupture of the aqueous film on the solid is
greater than that of the aqueous film on the oil.
Synergy between an oil which satisfies condition 52 and a particle which
satisfies condition 54 requires that the air-water contact angle conditions for
the particle to rupture air-water-air films be more severe than those for rup-
ture of air-water-oil films. This situation can, however, only occur for the
particle considered in Table I 0 if the configuration j = 4 occurs because
the conditions for 0Aw for all other configurations which yield rupture of
air-water-oil films would also mean rupture of air-water-air films (see Table
1). It is clear from Table l 0 that the configuration j = 4 can coexist with
both j = 5 and 6, where 00 w lies between 135° and 150°. In these circum-
stances the work of emergence of the particle into the oil-water surface from
the oil phase will largely determine the relative probabilities of the relevant
configurations (ignoring the effect of "degeneracy"). The work of emer-
gence WJ of each of the configurations at the oil-water surface is given (by
analogy with the corresponding Eq. 33 for the air-water surface)
wf = 'Yow(4lAsw cos(l80° - Bow) - 4lAow) (57)
where 4lAsw and 4lA 0 w are respectively the changes in solid-water and oil-
water surface areas accompanying emergence of the particles into the oil-
water surface from the oil phase. If we consider the emergence of a particle
into the oil-water surface of an oil droplet where the radius of the oil droplet
is large compared to the size of the particle, then we may neglect the re-
duction in surface area of the oil droplet due to changes in radius as the
particle emerges into the oil-water surface. The geometry of the problem of
estimating WJ then becomes exactly the same as that involved in estimating
the work W; of emergence of the same particle from the aqueous phase into
the air-water surface (where 180° - 00 w is substituted for 0Aw and 'Yow for
'YAw in the relevant expressions given in Table 1).
Whether configuration j = 4 has the largest negative work of emergence
depends upon 0ow· Thus, it is easy to deduce from the data presented in
Mode of Action of Antifoams 105

Table 10 that synergy may occur for these hypothetical particles if 00 w =


145° and 15° < 0Aw < 30°. This follows because under those conditions
configuration j = 4 is the most probable configuration (because it has the
largest negative work of emergence when coexistence with other configu-
rations is possible) and the particle cannot function as an antifoam alone
(because this requires 0Aw > 30°; see Table 1). In the case of 00 w = 140°,
however, configuration j = 5 at the oil-water surface is most probable and
collapse of the air-water-oil film will require 0Aw > 30°. This latter is, how-
ever, the same condition as that required for the particle to collapse air-
water-air films, so synergy will be absent in this case. Clearly, then, for
this hypothetical particle we have potentially complex behavior where syn-
ergy may or may not occur, depending upon the magnitudes of 00 w and
0Aw• which will in turn be influenced by the type and concentration of sur-
factant present.
No experimental study of the antifoam effect of mixing hydrophobed
crystals of well-characterized habit with oils has been reported. Experimen-
tal confirmation of the expectation of synergy only under restricted circum-
stances is therefore not available. However, Frye and Berg [39] have shown
evidence of only weak antifoam synergy when hydrophobed ground-glass
particles, which possess sharp edges (and are therefore relatively effective
antifoams when used alone), are mixed with hexadecane.
4. Rough Particles with Many Edges
The case of rough particles with many edges can be modeled by selecting
the axially symmetrical particle with regular rugosities shown in Fig. 21. In
the interests of clarity the particle is again depicted in Fig. 37 for the case
of interaction with an air-water-oil film. At the oil-water surface such a
particle may adopt configurations jj = 1 to 15 where that surface hinges at
the rugosities provided

0, < 0ow < 180° - 0, (58)

where 0, < 90° and is defined in Fig. 37. This particle can only cause rupture
of oil-water-air films when 0Aw < 90° if configuration jj = 8 is adopted
(where most of the particle is immersed in the oil) and 0Aw > e,. A particle
with such a configuration would also cause rupture of oil-water-oil films and
would therefore cause inversion of emulsion behavior so that water-in-oil
emulsions would be found.
The work of emergence from the oil phase wJ;o of each of the configu-
rations of this rough particle at the oil-water surface may be calculated in a
similar manner to WJ by using Eq. 57 if due allowance is made for the
relevant geometry. If again we assume that the particles are small relative
TABLE 10 Air-Water-Oil Film Collapse and the Configurations at the Oil-Water Surface of a Particle with Several Edges
Work of emergence• of particle
Air-water-oil into oil-water surface from oil =
Possible configurations at Conditions for re- film
Wf howX2 for Z = 5X
oil-water surface if 90° alization of collapse if SAw
configurations "Degeneracy" < 90° Sow = 145° Sow = 140°
<Sow< 180°

j = 1 150° < Sow< 180° 2


water
....
0
en nwnC/tmf~il

j=2 150° < Sow< 180° 4

j = 3
~ 90° < flow < 120° 2 Yes provided

v
()AW> 60°
=c
j=4 135° < 0ow < 165° 4 Yes provided -1.47 -2.00
0Aw > 15°

j = 5 30° < 0ow < 150° l Yes provided -1.37 -2.17


0AW > 30°

~
j=6 90° < 00 w < 180° 2 -0.33 -0.44
..... lfllilll'll""'
.....
Q

ill
Particle geometry as shown in Fig. 19 with 0P = 30° and Y = X. Z is the length of the particle defined in the figure.
'Calculated using expressions shown in Table l for emergence into the air-water surface from water, which is exactly analogous except that "Yow replaces
'YAw and 180° - 00 w replaces 0Aw·
108 Garrett

cp
'
ii= 1
ii= 15
ii= 2
ii= 14
ii =3
ii= 13
ii= 4 Oil-water interface will
ii= 12 hinge on rugosities jj = 1 ---> 15
jj = 5 if
ii= 11 e, <8ow <180° -8,
jj=6
ii= 10
ii= 7
ii= 9
ii= 8

x >I

(a) air

(b) air

~ ..... ~ flli'
oil l__j _,... L_j ..,. rupture

FIG. 37 Axially symmetrical particle with many edges at oil-water interface: (a)
example of configuration which stabilizes air-water-oil film where oil-water inter-
face hinges on any rugosity jj ~ 8; (b) configuration where oil-water interface hinges
on rugosity jj = 8 will cause air-water-oil film rupture provided tlAw > 8,.
Mode of Action of Antifoams 109

TABLE 11 Work of Emergence into the Oil-Water Surface wj;o of an Axially


Symmetrical Particle with Regular Rugosities

Configuration with oil-water surface


hinging on rugosity jj
0ow = 135° 0ow = 95°
Ji= 1 6.951 -0.391
2 5.908 -0.520
3 4.866 -0.648
4 3.823 -0.777
5 2.781 -0.905
6 1.738 -1.034
7 0.695 -1.163
8 -0.347 -1.291
0, = 30°; and H /X = 0.1; see Fig. 37 for definition of 0., H, and X.

to the size of the oil droplets, then we find

w:t-
~
'YowX
0
= 1T [ pjj (H/X)
-.--
sm 0,
(1 + -tanH/X)
- + 0.25 Jcos (180 - 0 )
0,
0w

[0.5 + -tanH/XJ
(59)
2
- 1T -
0,

where X is defined in Fig. 37, and Pii is the number of conical segments
of height H (see Fig. 37) transferred from the oil phase to the aqueous
phase.
Calculations of wi;o howX2 for a particle which satisfies both conditions
54 and 58, so that

90° < 00 w < 180° - 0, (60)

are given in Table 11. From this table it is clear that the configuration jj =
8, with the particle mainly in the oil phase, yields the largest negative value
of wi; 0 • That is therefore the most probable configuration.
An alternative configuration may occur if the particle is completely
removed from the bulk of the oil so that the oil just fills the rugosities.
This is illustrated in Fig. 38. It is the configuration considered by Dippenaar
[52], where the effective contact angle of the oil-filled particle at the
air-water surface is supposed higher than that of the uncontaminated par-
ticle. This configuration can occur if the oil-water contact angle satisfies the
condition
110 Garrett

water

w,•o 1
90° -8,

FIG. 38 Formation of hydrophobic particle with oil-filled rugosities after dispersal


of oil-particle mixture in aqueous phase.

90° + 0, < 00 w < 180° (61)

Conditions 60 and 61 overlap so that the configuration depicted in Fig. 38


can coexist with the configurations jj = 1 to 15 if

90° + 0, < 00 w < 180° - 0, (62)

However, the configuration where the rugosities are just filled with oil (shown
in Fig. 38) will have a higher work of formation w;
0 than that of the most

probable configuration at the oil-water surface where jj = 8 and where most


of the particle is immersed in the oil. This will arise because of both the
greater oil-water surface area and the greater particle-water surface area.
Thus, if we assume the particles are small compared to any oil droplet in
which they are initially dispersed then the difference in work of formation
Mode of Action of Antifoams 111

of the two configurations is

~/X) (i + H/X) + 0.25]cos (180° -


[( sm 00 w)
0, tan 0,

+ [o.5 + H/X]2
tan 0,
(63)

+[ 14(~) (1+ ::!) J


so that we always have w~ 0 > wt 0 provided 0ow > 90° since 0, < 90°.
We therefore find for a rough particle, which satisfies conditions 54 and
58 that the most probable configuration is one where the particle is mostly
immersed in the oil phase. The oil-water surface hinges on an edge so that
exposure of particle surface to the aqueous phase is minimized. As we have
seen, this is also the only configuration which can cause rupture of the oil-
water-air film (provided condition 39 is also satisfied so that 0Aw > 0,). By
contrast, a particle of the same geometry which satisfies condition 39 and
for which 0Aw < 90° will have a low probability of rupturing air-water-air
foam films. This follows because the only configuration which such a par-
ticle may adopt at the air-water surface which can cause rupture has the
lowest probability of occurrence (see Sec. IV.C.). Therefore, a rough par-
ticle of geometry shown in Fig. 37 (or 21) has a high probability of rupturing
an oil-water-air film if 00 w > 90° and 0, < 0Aw < 90° but a low probability
of rupturing an air-water-air foam film. Again, then, we would expect syn-
ergy if such particles are mixed with oils of low 'Y Ao so that the bridging
condition 52 is satisfied. Since most particles used as antifoam promoters
are best considered to be rough with many edges, then the ubiquitous oc-
currence of such synergy is clearly consistent with the argument outlined
here. The overall process of antifoam action is depicted schematically in
Fig. 39 for the case of an oil for which S" < 0 and 0* > 90° (i.e., condition
52).

F. Antifoam Dimensions and Kinetics


There have been no systematic studies of the effect of particle size on the
efficiency of particle-oil antifoams. However, if part of the role of the par-
ticle is to penetrate any electrical double layer between oil droplets and air
bubbles, then clearly the particle size should be at least of the same order
as the double layer thickness, i.e., >0.1 µm. If the particle size is appre-
ciably smaller than this, then the oil droplet will in effect be smooth so that
the presence of particles will be irrelevant (except in that they modify the
overall interaction forces between bubbles and antifoam entities).
112 Garrett

/
/
(a)

air

/
water~-

l
-
(b)

air

FIG. 39 Proposed overall process of foam film rupture by hydrophobic particle-


oil mixtures for oils where S" < 0 and 6* >90° (i.e., condition 52). (a) Particle
ruptures air-water-oil film permitting oil droplet to adhere to bubble surface (b) Oil
lens formed. (c) Oil lens bridges air-water-air foam film. (d) Capillary pressure in
vicinity of bridging lens increases rate of drainage of foam film. (e) Film rupture.
Mode of Action of Antifoams 113

Particle-oil antifoams form composite entities. Increase in particle size at


constant weight fraction of necessity increases the size of those entities,
otherwise the system will subdivide into oil droplets and oil-coated particles.
If these composites are to cause foam film rupture, then they must first form
a lens at the air-water surface at one side of the foam film and subsequently
emerge into the other side to bridge the film. Drainage of the foam film to
dimensions of the same order as that of the antifoam may therefore be ex-
pected to play the same role with oil-particle antifoams as with particle an-
tifoams. We would therefore expect that the considerations of Dippenaar
[52] and Frye and Berg [63] to be relevant for particle-oil antifoams so that
foam film draining is the rate-determining step for antifoam action. Sizes
of the antifoam particle-oil composites should therefore be small enough to
ensure a high probability of presence in a given foam film but not so small
that foam film drainage to the dimensions of the antifoam entity is too slow.
Observations with both mineral oil/hydrophobed silica [40] and silicone [127]
antifoams suggest that improvements in antifoam effectiveness of a given
weight concentration of antifoam may be achieved by decreasing antifoam
entity sizes down to at least 1-2 µm.

ACKNOWLEDGMENTS
The author wishes to acknowledge many discussions with Prof. D.G. Hall
concerning the calculation of the probabilities of particles adopting various
configurations at the fluid-fluid surface. The author also wishes to thank Dr.
R. Aveyard and Dr. J. Lucassen for reading through the manuscript and
making helpful suggestions.

REFERENCES
l. J. A. Kitchener in Recent Progress in Surface Science (J. F. Danielli, K. G.
A. Pankhurst, and A. C. Riddiford, eds.), Vol. 1, Academic Press, New
York, 1964, p.51.
2. J. Lucassen in Anionic Surfactants-Physical Chemistry of Surfactant Action
(E. H. Lucassen-Reijnders, ed.), Marcel Dekker, New York, 1981, p.217.
3. A. Prins in Advances in Food Emulsions and Foams (E. Dickinson, G. Stainsby,
eds.), Elsevier Applied Science, New York, 1988, p.91.
4. P. Walstra in Foams: Physics, Chemistry and Structure A.J. Wilson, ed.),
Springer-Verlag, Berlin, 1989, p. l.
5. R. S. Hansen in Foams (R. J. Akers, ed.), Academic Press, London, 1976,
p.l.
6. J. W. Gibbs in The Scientific Papers, Vol.l, Dover, New York, 1961.
7. K. Malysa, R. Miller, and K. Lunkenheimer, Colloids Surfaces 53: 47 (1991).
114 Garrett

8. A. Prins, In Foams (R. J. Akers, ed.), Academic Press, London, 1976, p.51.
9. A. Vrij and J. Th. G. Overbeek, J. Am. Chem. Soc. 90: 3074 (1968).
10. B. P. Radoev, A. D. Scheludko, and E.D. Manev, J. Colloid Interface Sci.
95: 254 (1983).
11. D.R. Exerowa, lzv. Khim 11(3-4): 739 (1978).
12. E. D. Goddard and G.C. Benson, Can. J. Chem. 35:986 (1957).
13. J. M. Corkill, J.F. Goodman, and S.P. Harrold, Trans. Faraday Soc. 60:
202 (1964).
14. C. H. Fiske, J. Biol. Chem. 35: 411 (1918).
15. T. Sasaki, Bull Chem. Soc. Japan 11: 797 (1936).
16. T. Sasaki, Bull Chem. Soc. Japan 13: 517 (1938).
17. L. T. Shearer and W.W. Akers, J. Phys. Chem. 62: 1264, 1269 (1958).
18. J. W. McBain, S. Ross, A. P. Brady, J. V. Robinson, I. M. Abrams, R.
C. Thorburn, and C. G. Lindquist, National Advisory Committee for Aer-
onautics A.R.R. No. 4105, 1944.
19. J. V. Robinson and W. W. Woods, J. Soc. Chem. Ind. 67: 361 (1948).
20. S. Ross, J. Phys. Colloid Chem. 54: 429 (1950).
21. S. Ross, R. M. Haak, J. Phys. Chem. 62: 1260 (1958).
22. S. Okazaki, K. Hayashi, and T. Sasaki, Proceedings of the IV International
Congress on SAS, V3, Brussels, 67, 1964.
23. T. A. Koretskaya and P. M. Kruglyakov, Izv. Sib. Otd. An SSSR, Ser. Khim.
Nauk. 7(3): 129 (1976).
24. S. Ross and T. H. Bramfitt, J. Phys. Chem. 61: 1261 (1957).
25. E. J. Burcik, J. Colloid Sci. 5: 421 (1950).
26. W. E. Ewers and K. L. Sutherland, Aust. J. Sci. Res. 5: 697 (1952).
27. J. A. Kitchener and C. F. Cooper, Quart. Rev. 13: 71 (1959).
28. A. Leviton and A. Leighton, J. Dairy Sci. 18: 105 (1935).
29. S. Ross and G. J. Young, Ind. Eng. Chem. 43 (11): 2520 (1951).
30. R. D. Kulkarni and E. D. Goddard, Croatica Chem. Acta 50(1-4): 163 (1977).
31. P. M. Kruglyakov and P.R. Taube, Zh. Prikl. Khim. 44(1): 129 (1971).
32. W. D. Harkins, J. Chem. Phys. 9: 552 (1941).
33. J. S. Rowlinson and B. Widom in Molecular Theory of Capillarity, Oxford
University Press, Oxford, 1982.
34. P. M. Kruglyakov, in Thin Liquid Films, Fundamentals and Applications (I.
B. Ivanov, ed.), Marcel Dekker, New York, 1988, p. 767.
35. P. M. Kruglyakov and T. A. Koretskaya, Kolloid. Zh. 36(4): 682 (1974).
36. A. P. Koretsky, A. V. Smirnova, T. A. Koretskaya, and P. M. Kruglyakov,
Zh. Prikl. Khim. 50: 84 (1977).
37. T. A. Koretskaya, Kolloid. Zh. 39(3): 571 (1977).
38. M. N. Fineman, G. L. Brown, and R. J. Myers J. Phys. Chem. 56: 963
(1952).
39. G. C. Frye and J.C. Berg, J. Colloid Interface Sci. 130(1): 54 (1989).
40. P. R. Garrett, J. Davis, and H. M. Rendall, unpublished work.
41. Y. Abe and S. Matsumura, Tenside Detergents 20(5): 218 (1983).
42. I. C. Callaghan, C. M. Gould, R. J. Hamilton, and E. L. Neustadter, Col-
loids Surfaces 8: 17 (1983).
Mode of Action of Antifoams 115

43. C. B. McKendrick, S. J. Smith, and P. A. Stevenson, Colloids Surfaces 52:


47 (1991).
44. P. R. Garrett, J. Colloid Interface Sci. 69(1): 107 (1979).
45. A. Prins in Food Emulsions and Foams (E. Dickinson, ed.), Royal Society
of Chemistry Special Publication 58, 1986, p.30.
46. J. A. Fay in Oil on the Sea (D. P. Hoult, ed.), Plenum Press, New York,
1969, p.53.
47. D. P. Hoult, Ann. Rev. Fluid Mech. 4: 341 (1972).
48. C. Huh, M. Inoue, and S. G. Mason, Can. J. Chem. Eng. 53: 367 (1975).
49. P. Joos and J. Pintens, J. Colloid Interface Sci. 60: 507 (1977).
50. S. U. Pickering, J. Chem. Soc., 2001 (1907).
51. 0. Bartsch, Kolloidchem. Beihefte 20: 1 (1924).
52. A. Dippenaar, lnt. J. Mineral Process. 9: 1-22 (1982).
53. V. M. Lovell in Flotation; A. M. Gaudin Memorial Volume (M. C. Feur-
stenau, ed.), Vol. 1, American Institute of Mining, Metallurgical, and Pe-
troleum Engineers, Inc., 1976, p. 597.
54. J. R. Tate and A. C. McRitchie (assigned to Procter and Gamble), GB
1,492,938, November 23, 1977, filed January l, 1974.
55. S. G. Mokrushin, Kolloidn. Zh. 12: 448 (1950).
56. N. Dombrowski and R. P. Fraser, Phil. Trans. Royal Soc. London, Ser.A
247: 13 (1954).
57. A. K. Livshitz and S. V. Dudenkov, Tsvet. Metally 30(1): 14 (1954).
58. J. H. Schulman and J. Leja, Trans. Faraday Soc. 50: 598 (1954).
59. A. K. Livshitz and S. V. Dudenkov, Proceedings of 7th IMPC New York,
(1965), p. 367.
60. S. V. Dudenkov, Tsvet. Metally 40: 18 (1967).
61. R.R. Irani and C. F. Callis, J. Phys. Chem. 64: 1741 (1960).
62. H. Peper, J. Colloid Sci. 13: 199 (1958).
63. G. C. Frye and J.C. Berg, J. Colloid Interface Sci. 127(1): 222 (1989).
64. A. J. De Vries, Ree. Trav. Chim. 77: 383 (1958).
65. A. K. Livschitz and S. V. Dudenkov, Tsvet. Metally 33: 24 (1960).
66. F. Tang, A. Xiao, J. Tang, and L. Jiang, J. Colloid Interface Sci. 131(2):
498 (1989).
67. J. F. Oliver, C. Huh, and S. G. Mason; J. Colloid Interface Sci. 59(3): 568
(1977).
68. A. J. Hopkins, and L. V. Woodcock, J. Chem. Soc. Faraday Trans. 86(12):
2121 (1990).
69. R. Vacher, T. Woignier, J. Pelous, and E. Courtens, Phys. Rev. B. 37(11):
6500 (1988).
70. K. J. Mysels, K. Shinoda, and S. Frankel in Soap Films-Studies of Their
Thinning, Pergamon Press, London 1959.
71. 0. Reynolds, Phil. Trans. Royal Soc. London, Ser. A 177: 157 (1886).
72. A. Vrij, Disc. Faraday Soc. 42: 23 (1966).
73. T. D. Blake and J. A. Kitchener, J. Chem. Soc. Faraday Trans. 1, 68: 1435
(1972).
116 Garrett

74. J. N. Israelachvili in Intermolecular and Surface Forces with Applications to


Colloidal and Biological Systems, Academic Press, London, 1985.
75. J. F. Anfruns and J. A. Kitchener, Trans Inst. Min. Metall., London C 86:
9 (1977).
76. W. A. Ducker, R. M. Pashley and B. Ninham, J. Colloid Interface Sci.
128(1), 66 (1989).
77. J. Strnad, H. Kohler, K. Heckmann, and M. Pitsch, J. Colloid Interface Sci.
132(1): 283 (1989).
78. B. Derjaguin, Trans. Faraday Soc. 36: 203 (1940).
79. J. Davis and P. R. Garrett, unpublished work.
80. M. Aronson, Langmuir 2, 653 (1986).
81. R. D. Kulkarni, E. D. Goddard, and B. Kanner, Ind. Eng. Chem., Fundam.
/6(4): 472 (1977).
82. R. Birtley, J. Burton, D. Kellett, B. Oswald, and J. Pennington, J. Pharm.
Pharmac. 25: 859 (1973).
83. M. J. Povich, A. I. Ch. E. J. 21(5): 1016 (1975).
84. S. Ross and G. Nishioka in Emulsions, Lattices and Dispersions (P. Becker
and M. Yudenfreud, eds.), Marcel Dekker, New York, 1978, p. 237.
85. S. Ross and G. Nishioka, J. Colloid Interface Sci. 65(2): 216 (1978).
86. M. N. A. Carter and P.R. Garrett (assigned to Unilever Ltd.), GB 1,571,502,
July 16, 1980, filed January 23, 1976.
87. 0. F. Schweigl and G. P. Best (assigned to Unilever Ltd.), GB 1,099,502,
January 17, 1968, filed July 8, 1965.
88. P. R. Garrett (assigned to Unilever Ltd.), EP 75,433, March 30, 1983, filed
September 16, 1981.
89. S. Ross, Chem. Eng. Prog. 63(9): 41 (1967).
90. C. C. Currie (assigned to Dow Chemical Co.), US 2,632,736; March 24,
1953, filed August 22, 1946.
91. M. M. Solomon (assigned to General Electric Co.), US 2,829,112; Aprill,
1958, filed September 22, 1955.
92. S. Nitzsche and E. Pirson (assigned to Wacker-Chemie GmbH), US 3,235,509;
February 15, 1966, filed October 3, 1962.
93. (Assigned to Degussa), Fr 1,533,825; July 19, 1968, filed August 8, 1967.
94. M. J. O'Hara and D. R. Rink (assigned to Union Carbide Corp.), GB
1,247,690, September 29, 1971, filed August 11, 1987.
95. K. W. Farminer and C. M. Brooke (assigned to Dow Corning Ltd.), US
3,843,558; October 22, 1974, filed June 16, 1972.
96. G. Bartolotta, N. T. de Oude, and A. A. Gunkel (assigned to Procter and
Gamble Co.), GB 1,407, 997; October l , 1975, filed August 1, 1972.
97. F. J. Boylan (assigned to Hercules Powder Co.), US 3,076,768; February 5,
1963, filed Apri 5, 1960.
98. R. Leibling and N. M. Canaris (assigned to Nopco Chemical Co.), US
3,207,698, September 21, 1965, filed February 13, 1963.
99. R. E. Sullivan (assigned to Dow Corning Corp.), US 3,304,266, February
14, 1967, filed May 6, 1963.
Mode of Action of Antifoams 117

100. E. Domba (assigned to Nalco Chemical Co.), US 3,388,073, June 11, 1968,
filed December 16, 1963.
101. H. Buckman (assigned to Buckman Laboratories Ltd.), GB 1,166,877, Oc-
tober 15, 1969, filed December 8, 1967.
102. G. C. Harrison and A. J. Stumpo (assigned to Pennsalt Chemicals Corp.),
GB 1, 195 ,589; June 17, 1970, filed August 10, 1967.
103. H. Lieberman, C. A. Duharte-Francia, and J. W. Henderson (assigned to
Betz Laboratories Inc.), GB 1,267 ,479; March 23, 1972, filed May 26, 1969.
104. J. R. Miller, R. H. Pierce, R. W. Linton, and J. H. Wills (assigned to Phil-
adelphia Quartz Co.), US 3,714,068; January 30, 1973, filed December 28,
1970.
105. M. Caviet (assigned to Shell Development Co.), Canada 508,856; April 4,
1955, filed November 28, 1949.
106. F. J. Boylan (assigned to Hercules Inc.), US 3,705,859; December 12, 1972,
filed December 30, 1970.
107. G. Boehmke, M. Quaedrlieg, and G. Kolla (assigned to Farbenfabriken Bayer
Aktiengesellschaft), GB 1,267 ,482; March 22, 1972, filed July 14, 1970.
108. H.J. Shane, J. E. Schill and J. W. Lilley (assigned to Hart Chemical Ltd.),
Canada 922,456; March 13, 1973, filed March 31, 197 l.
109. I. A. Lichtman and A. M. Rosengart (assigned to Diamond Shamrock Corp.),
Canada 927,707; June 5, 1973, filed September 3, 1971.
110. (Assigned to Henkel and Cie GmbH), GB 1,386,042; March 5, 1975, filed
January 25, 1973.
111. F. M. Ernst (assigned to Mobil Oil Corp.), US 3,909,445; September 30,
1975, filed September 21, 1972.
112. F. J. Boylan (assigned to Hercules Inc.), US 3,408,306; October 29, 1968,
filed July 7, 1961.
113. A. G. Kanellopoulos and M. J. Owen, Trans. Faraday Soc. 67: 3127 (1971).
114. P. R. Garrett and P. Gratton, unpublished work.
115. J. Timmermans, Physico-Chemical Constants of Pure Organic Compounds,
Vols. l, 2, Elsevier, Amsterdam, 1950.
116. R. Aveyard and D. A. Haydon, Trans. Faraday Soc. 61: 2255 (1965).
117. R. Aveyard, P. Cooper, and P. D. I. Fletcher, J. Chem. Soc. Faraday Trans.
86(1): 21 I (1990).
118. R. Aveyard, B. P. Binks, P. Cooper, and P. D. I. Fletcher, Adv. Colloid
Interface Sci. 33: 59 (1990).
119. F. Van Voorst Vader, Chem. lng. Techn. 49(6): 488 (1977).
120. A. A. Trapeznikov and L. V. Chasovnikova, Colloid J. USSR 35: 926 (1973).
121. P.R. Garrett, J. Colloid interface Sci. 76(2): 587 (1980).
122. P. R. Garrett and P. Moore, unpublished work.
123. J. Sinka and I. Lichtman, lnt. Dyer Textile Printer: (May), 489 (1976).
124. R. D. Kulkarni, E. D. Goddard, and B. Kanner, J. Colloid Interface Sci.
59(3): 468 (1977).
125. M. A. J. S. van Boeke! and P. Walstra, Colloids Surfaces 3: 109 (1981).
126. J. Mizrahi and E. Barnea, Br. Chem. Eng. 15(4): 497 (1970).
127. V. Veber and M. Paucek, Acta Fae. Pharm. Univ. Comenianae 26: 221
(1974).
2
Antifoams for Nonaqueous Systems in
the Oil Industry

IAN C. CALLAGHAN BP Research, Sunbury-on-Thames, Middlesex,


United Kingdom

I. Introduction 119
II. Occurrence of Foams in the Oil Industry 121
A. Crude oil production 121
B . Oil refining and oil products 125
III. Characterization of Nonaqueous Foam Systems 127
A. Foam structures 128
B. Methods of assessing the foaming characteristics
of nonaqueous systems 128
IV. The Relationship between Physical Properties and
Foam Stability 138
A. Bulk rheological properties 138
B. Nature of the filler gas 139
C. Composition of the foaming system 140
D. Interfacial rheological properties 140
V. Antifoam Types and Methods of Selection 143
A. Antifoam types 143
B. Selection of antifoams 146
VI. Future Directions 147
Acknowledgments 147
References 147

I. INTRODUCTION
The stability and breaking of nonaqueous foams is a subject of great im-
portance to the oil industry, as such foams occur, for example, in the pro-
duction and refining of crude oil, as well as during the use cycle of many
119
120 Callaghan

products such as lubricants. Foam, whenever it occurs, leads to a reduction


in process/product efficiency with consequent loss of revenue, if the prob-
lem remains untreated. Consequently, the reliable assessment of the foam-
ability of oils and oil products is of prime importance as it allows potential
foaming problems to be anticipated and allowed for at an early stage in either
the design of a production/refining process or in the development of an oil
product. However, despite the obvious importance of such nonaqueous foams,
much of the available literature on foam stability and breaking relates to
aqueous systems, with no clear distinction being made between them and
nonaqueous systems. Indeed, Ross [l] having reviewed aqueous foams in
some detail, states that "the same principles can be applied mutatis mutandis
to the problem of foam in non-aqueous media."
A number of authors [2-11] have worked specifically on nonaqueous
foams and reported that foam stability is due to a number of factors such
as bulk viscosity, surface rheology, and the presence of surface active spe-
cies. Brady and Ross [2] studied a series of engine oils and medicinal-grade
paraffin oils and showed that foam stability increased linearly with the ki-
nematic viscosity of the oil. McBain and Robinson [3] confirmed this view
and showed that a high viscosity, both in the bulk and at the surface, was
often associated with high foam stability, but cautioned that viscosity alone
was not the sole cause. They attributed the high surface viscosities they had
observed to the formation of plastic films at the gas-liquid interface giving
rise to a semi-immobilized oil layer underneath. Ross [1] reported that in
aqueous foams x-ray diffraction studies had shown that such structures did
indeed exist. He has also described experimental evidence for the existence
of surface plasticity in crude and fuel oils [5]. However, in contrast, Man-
nheimer and Schecter [6] did not find any evidence for surface viscosity
being an important factor in determining the stability of mineral oil/calcium
sulphonate foam systems. Furthermore, Scheludko and Manev [7] have shown
that the thickness at which foam films rupture occurs spontaneously in the
aniline/lauryl alcohol system is independent of viscosity.
For crude oils, the presence of specific acids and phenols has been shown
[8] to be important for foam formation and stability. The presence of surface
active materials has also been shown by other authors [4,9, 10, 12] to be
necessary for foaming to occur in nonaqueous media. However, although
surface tension [7] has been shown to play a major role in determining non-
aqueous foam stability, it does not appear to be as important as the preceding
factors [3,11]. Dynamic surface tension effects, on the other hand, have
been postulated [13] as being particularly important to foam stability. Al-
though a number of such studies, e.g., [14,15], have been reported for aqueous
systems, only a relatively small number [8, 16, 17] have been reported for
nonaqueous systems.
Antifoams in the Oil Industry 121

The main method recommended for breaking nonaqueous foams is the


use of foam inhibitors (antifoams). Normally, these are added to the foaming
system in a suitable diluent. It has been shown experimentally [9] that many
effective antifoams have low solubility in the foaming medium. This was
confirmed by Robinson and Woods [18], who also studied the effects of the
surface tensions of the foaming medium and antifoam, as well as that of the
interfacial tension between them. They concluded that the main requirement
for an antifoam was that a droplet of the antifoam should penetrate the gas-
liquid interface of the foam on contact. They defined a rupture or entering
coefficient, E, as follows:
E = 'YF + 'YoF - 'Yo (1)

where
'YF = surface tension of the foaming medium
'Yo = surface tension of the antifoam
'YoF = interfacial tension between the antifoam and foaming medium

Using this parameter, they argued that for foam breaking to occur E must
have a high positive value. Ross [ l, 19] considered that the main requirement
for an antifoam was its ability to spread over the foaming liquid. He sug-
gested using Harkins' [20] spreading coefficient, S, to characterize foam
inhibitors:
S = 'YF - 'YDF - 'Yo (2)
In either case, it appeared that an effective antifoam must be readily dis-
persible in the foaming system, and must have a lower surfac~ tension than
that of the foaming liquid. This latter point was confirmed by Shearer and
Akers [21] in a study on the use of silicone oils as foam inhibitors for lu-
bricating oils. These authors showed that silicone oils were only effective
as antifoams if they were dispersed as droplets of less than 100 µm in diameter.
In this chapter, many of the points brought out above will be highlighted
further when we discuss the specific problems encountered with foams in
the oil industry.

II. OCCURRENCE OF FOAMS IN THE OIL


INDUSTRY
A. Crude Oil Production
When crude oil is produced, it is forced from the reservoir by pressure up
the production tubing to the surface (see Fig. 1). As the oil rises up the
tubing, the pressure becomes less and the associated gas is progressively
122 Callaghan

INJECTION PRODUCTION
WELL WELL
CAP ROCK

j j

FIG. 1 Production of oil from an oil reservoir.


Antifoams in the Oil Industry 123

mesh type
demister
vane type
demister
surge baffle

de-gassing element

water boot

FIG. 2 A gas-oil separator. (From Ref. 24, copyright © 1985, SPE-AIME.)

released from solution. On emerging from the well, it is usually necessary


to process the oil-gas mixture to remove both free and dissolved gas, which
is likely to come out of solution when the oil is maintained at about at-
mospheric pressure, as, for example, during transport in a tanker or storage.
Sometimes the separation of the gas is effected near the wellhead or, alter-
natively, the "live" crude oil may be conveyed under high pressure, e.g.,
30 bar, in a pipeline to a distant location whereupon the separation of the
oil and gas is effected. Occasionally, during this process, the crude oil forms
a stable foam, with the result that liquid oil is carried over into the gas
stream, creating serious problems in downstream compression plant. Such
carryover situations can, if not treated, result in severe production losses
and long equipment downtimes. To overcome this type of problem various
procedures can be used: they may be totally physical in nature or a com-
bination of both physical and chemical methods.
1. Physical Methods of Foam Control
It is common to describe the device used to separate gas from oil as a gas-
oil or three-phase separator (see Fig. 2) when the bulk component and prin-
cipal product are liquid hydrocarbon, at a nominally "steady" production
124 Callaghan

rate. The term three-phase separator is used in this context since water is
also separated from the gas and oil phases during this process.
The factors which affect separator design and performance are as follows:
a. the properties of the reservoir fluids, such as fluid densities and com-
positions, solids loading, and viscosities
b. the process conditions upstream of the separator
c. the process conditions downstream
d. external influences such as that of imposed motion, as for example, in
floating production systems

The liquid-handling capacity of a separator is generally determined by a


nominal mean residence time, typically 2 to 5 min. The selection of this
residence time is based partly on an assessment of foam lifetime (gas dis-
engagement time), partly on surge control requirements. The assessment of
gas disengagement time has been based, in the past, on design experience
and not on quantitative data. However, much effort is now being expended
on improving this situation through the modeling of foam collapse under
separator conditions, so as to allow separators to be designed with specific
crudes in mind. The recent improvements in the laboratory assessment of
foam characteristics under simulated separator conditions is proving useful
in this latter respect.
In many gas-oil separators, internal devices are employed to help improve
performance. The most common types of internal device are outlined briefly
below.
(a) Transverse Baffies (see Fig. 2). These are located perpendicular to
the direction of flow and act as skimmers to hold back the foam in the upper
part of the separator and, thereby, allow time for collapse to occur.
(b) Parallel Plates. These devices are very common and are variously
known as Dixon (Natco) or Arch (B.S.&B.) plates. The basic principle on
which these plates operate is that the separating fluids are forced to flow in
thin layers, giving rise to increased separation area, and some mechanical
enhancement of foam breaking. It is believed that the orientation of these
plates has some effect on their efficiency. However, as the bulk flow of
fluid is normally along the separator axis, this claim is by no means obvious.
There is some evidence from the author's studies that the wetting charac-
teristics of the plate materials used to manufacture these devices may have
some bearing on their efficacy as foam breakers.
(c) Random Packings. These are similar to those used in mass transfer
columns, and they can assist in the breaking of crude oil foams or in pre-
venting their formation. Again the wetting characteristics of the materials
used to make these packings has an influence on their overall performance.
Antifoams in the Oil Industry 125

(d) Inlet Devices. These may be cyclonic, dished, or flat-plate construc-


tions. They are claimed to assist gas-liquid separation, but may in fact pro-
mote foam formation and hence worsen the overall situation in some instances.
( e) Wire Mesh Pads. These are commonly installed at the gas outlet of
a separator to act as demisters, removing any entrained droplets from the
gas stream. Special meshes of lower bulk density are used to promote foam
collapse. These latter devices are positioned in the separator headspace, up-
stream of the demister pads, and are thought to function by stressing the
foam bubbles as they pass through the mesh, causing them to rupture.

2. Chemical Methods of Foam Control


The use of chemicals to prevent foam formation on entry into a gas-oil sep-
arator is well established. The principal types of antifoam used in this ap-
plication are polydimethylsiloxanes and, for particularly difficult or aggres-
sive crude oils, fluorosilicones. These liquids are ideally added sufficiently
far upstream of the separator to allow thorough dispersal throughout the
crude oil: this process can be aided by the proper choice of carrier solvent.
In general, the dose rates required for these two classes of chemical fall
within the range of 0.1 to 20 ppm. It is important to ensure that the correct
dosage is predete1mined in the laboratory, since profoaming can occur if
the antifoam is applied (in the field) at a concentration below its solubility
limit.

B. Oil Refining and Oil Products


1. Crude Oil Distillation Units
Foaming in crude oil distillation columns can lead to flooding of the trays,
causing foam to pass down the downcomers (see Fig. 3), resulting in a loss
of separation efficiency due to the recycling of vapour. It has been suggested
[22) that this problem can be solved by careful tray design to produce jetting
rather than bubbling at the perforations. Alternatively, one can use antifoam
chemicals to reduce the foaming tendency of the liquid being distilled: the
preferred materials are again the high molecular weight polydimethylsiloxanes.

2. Thermal Cracking Processes


When hydrocarbons are heated to temperatures in excess of 450°C they be-
gin to decompose or "crack" into smaller molecular units. Paraffinic hy-
drocarbons are the most easily cracked materials followed by naphthenes,
aromatic hydrocarbons being extremely refractory. As more heavy feed-
stocks have become available, the importance of thermal cracking processes
has increased. One such process which is currently in use is visbreaking.
Visbreaking is the thermal cracking of viscous residues to reduce their vis-
126 Callaghan

DOWNCOMER TRAY ABOVE

LIQUID
--- -/- --
FOAM

TRAY BELOW

FIG. 3 Foam in a distillation column downcomer.

cosity by breaking down the large complex molecules in the feedstock to


smaller ones. Using this process a satisfactory fuel oil can be made with
minimum recourse to the blending of gas-oil or kerosine with the viscous
residue.
Another thermal cracking process is delayed coking. This process is nor-
mally applied to atmospheric or vacuum residues from low-sulfur crudes for
the production of electrode-grade coke. Also, it is increasingly being used
for the disposal of heavy residues from high-sulfur crudes. In both these
processes, the release of volatile products as the feed is heated can lead to
foam formation with a consequent reduction in capacity, and fouling of over-
head lines. The use of silicone-based antifoams controls the foam buildup
and allows faster feed rates, greater capacity, and a reduction in fouling of
the overhead lines. High molecular weight silicone fluids are again the pre-
ferred materials in these applications.

3. Lubricating Oils
Lubricating oils cover a diverse range of products, specially blended to meet
specific in-use requirements in such machinery as compressors, steam tur-
bines, and internal combustion engines. Lubricating oils are classified ac-
cording to end use and include automotive lubricants, metalworking fluids,
and compressor oils. The formation of foam during the use of these oils is
deleterious to their performance and needs to be avoided at all costs. This
problem is usually solved through careful formulation of the product and
involves the use of foam inhibitors such as alkyl polyacrylates and low mo-
lecular weight silicone oils. Of these two classes of antifoam, the alkyl
Antifoams in the Oil industry 127

polyacrylates are to be preferred since they have much better air-release


properties than silicone oils.

4. Automotive Fuels
The foaming of automotive fuels can be problematic, leading to such prob-
lems as difficulty in completely filling a car's fuel tank at a service station.
This problem is particularly pronounced with diesel fuel and can result in
overflow and early filler pump cutout. As in the case of lubricating oils, the
solution to the problem lies in proper formulation of the fuel's additive pack-
age. Silicone-polyether-type antifoams appear to be the preferred materials
for this application, although both dimethyl silicone oils and fluorosilicones
also have utility if formulated correctly.

5. Propane Deasphalting
During the removal of asphaltenes from residue by precipitation with deas-
phalting agents, foam may be produced in the tower mixer during flash dis-
tillation. This foaming causes crude oil concentrate to overflow with con-
sequent loss of process efficiency. The addition of a silicone antifoam to
the feed to the deasphalting tower prevents foaming and the escape of re-
sidual oil in the spray condenser and asphalt stripper. However, the main
potential problem with foam in this process is during the steam stripping of
the asphalt. Normal practice in this case would be to add a silicone antifoam
agent to the feed to the stripper rather than to the deasphalting tower.

111. CHARACTERIZATION OF NONAQUEOUS FOAM


SYSTEMS

Foams are colloidal systems in which a gaseous phase is dispersed through-


out a continuous liquid phase in such a way that the gas cavities are sepa-
rated by thin liquid films. It is the stability of these liquid films that deter-
mines the overall stability of the foam. A fundamental characteristic common
to all foams is their very large interfacial area. As a consequence, they are
not stable in the thermodynamic sense, although some aqueous foams, e.g.
shaving foam, can exist in a metastable state for very long periods in the
absence of any external disturbances. Nonaqueous foams, on the other hand,
are usually (though not always) incapable of existing in such a state and
undergo complete collapse on removal of the motivation to foam, e.g., re-
moving the source of filler gas. Thus, the stability of nonaqueous foams is
usually discussed in terms of their collapse kinetics: stable foams being those
that undergo collapse at a slow rate.
128 Callaghan

A. Foam Structures
There are two extreme structures which foams can assume; these are the
classical polyhedral structure and the round-bubbled or gas dispersion struc-
ture. Both structures can be experienced in nonaqueous foam systems.
Polyhedral foams comprise large pockets of gas separated by thin liquid
films. These liquid films form a three-dimensional network down which liq-
uid drains from the foam under the influence of capillary forces. A typical
crude oil polyhedral foam is shown in Fig. 4. Round-bubbled foams, on the
other hand, consist of a dispersion of nearly spherical gas bubbles separated
by relatively thick films of the bulk liquid. This type of foam has been
termed kugelschaum by Manegold [23]. An example of a round-bubbled
crude oil foam is shown in Fig. 5. In crude oil systems, both extremes of
foam can be found and the particular type encountered often depends on the
prevailing conditions of temperature and pressure under which the foam is
formed. Since foam type can influence the choice of antifoam employed, it
is clearly important to ascertain whether a round-bubbled or polyhedral foam
forms under a given set of process conditions. For example, round-bubbled
foam requires a silicone antifoam of low polydispersity [24], whereas poly-
hedral foam can be treated with a conventional polydisperse silicone oil. In
the case of crude oils, polyhedral foams tend to form in high gas-oil ratio
(GOR) systems at low pressures, whereas round-bubbled foams tend to form
in either high GOR systems at moderate to high pressures or in low GOR
systems. These latter structures eventually undergo syneresis to the poly-
hedral state as liquid drains from between the gas bubbles (see Fig. 6).

B. Methods of Assessing the Foaming


Characteristics of Nonaqueous Systems
1. General Methods
The simplest apparatus in which foam can be studied is that due to Bikerman
[25] (see Fig. 7). Bikerman's gas sparging tube comprises a glass column
with a glass sinter at one end onto which a known volume of the test liquid
is pipetted. The flow of gas through the sinter causes the liquid to foam:
different filler gases can be used to simulate different situations. Callaghan
and Neustadter [26], in a study of crude oil foaming, used a modified Bik-
erman foaming column that was 30 cm in length with two fine (no. 2 po-
rosity) sintered glass disks placed 1 cm apart, situated at the base of the
tube just above the gas inlet. The gas used to create the foam was admitted
to the column via a pressure reduction and flowmeter assembly after being
saturated with the crude oil by passage through a series of Dreschel bottles
Antifoams in the Oil Industry 129

FIG. 4 Polyhedral crude oil foam. (Reproduced with permission from A. J. Wil-
son, Foams: Physics, Chemistry, and Structure, Chapter 7, Springer-Verlag, London,
1989.)

containing the oil. These workers used a fixed gas flowrate of 40 mL/min
and allowed the aliquot of oil to be completely taken up into the foam before
extinguishing the gas supply and recording the decrease of foam height with
time. Using this method, two parameters which characterize the foam can
be derived: these are the foaminess index (!,) and the average foam lifetime
FIG. 5 Round-bubbled crude oil foam. (Reproduced with permission from A. J.
Wilson, Foams: Physics, Chemistry, and Structure, Chapter 7, Springer-Verlag,
London, 1989.)

130
Antifoams in the Oil Industry 131

0 0
0
0 0 Gas Diffusion
00 00 Liquid Drainage
0 0 0

Gas Dispersion Polyhedral Foam

0
Gas / ~
o-Bubbles
\ ~
Gas \ \v'
t:as
Bulk
Liquid 0
Pipeline Crude Stabilized Crude
24mN/m 30mN/m
Typical Surface Tension

FIG. 6 Foam syneresis. (From Ref. 24, copyright© 1985, SPE-AIME.)

(Lro). The foaminess index is the ratio of the foam volume generated by a
given gas flowrate to that flowrate:

foam volume
Foaminess index 4 = ----- (3)
gas flowrate

This parameter has dimensions of time: it is a measure of foam-forming


tendency. The average foam lifetime is determined from the collapse curve
(see Fig. 8) and is a measure of foam stability. It too has the dimensions
of time.

iTH(t) dt

Average foam lifetime LF = - - - - (4)


H(O)

where
H(O) = steady foam height at time t = 0
H(t) = foam height at time t
132 Callaghan

graduated glaBS cylinder

F
L
0
w
c
0 gas
N supply
T
R
0
L
L
E
R

Oil

\
Dreachel bottle

FIG. 7 Bikerman sparge tube apparatus.

T = time at which the foam height has reached zero; i.e., all foam
bubbles have disappeared
The choice between foaminess index and average foam lifetime as the
measured parameter is very much dependent upon the system under inves-
tigation. If that system is a gas-oil separator then the average lifetime is
generally measured, since it gives information that can be readily related to
residence time in the separator. However, where the system is a distillation
Antifoams in the Oil Industry 133

-'-f ~.dT
Time

LF
H(o) 0

Area under curve


Initial foam height

FIG. 8 Foam collapse curve (schematic). (Reproduced with permission from A. 1.


Wilson, Foams: Physics, Chemistry, and Structure, Chapter 7, Springer-Verlag,
London, 1989.)

column, the foaminess index is more appropriate since, in distillation pro-


cesses, the liquid (downcoming) phase is continuously sparged with the gas-
eous (rising) phase. In general, however, it is likely that the choice is ar-
bitrary since the two parameters are closely related.
2. Methods Specific to Crude Oil Systems
A major disadvantage of the Bikerman sparge tube method as applied to
crude oil foaming is that it does not reflect the real situation in, say, a gas-
oil separator. Consequently, a method of assessing foam under high pres-
sure/temperature conditions is required. Such an apparatus is shown [27] in
Fig. 9. The foam is created by allowing a sample of "live" crude oil to flow
from an injection vessel at, say, pipeline pressure into a sample cell held at
gas-oil separator pressure and temperature. This sudden depressurisation causes
the crude oil to degas and form a foam. Once the foam has been fully formed
the sample is isolated from the reference cell, and the subsequent foam col-
lapse monitored by a differential pressure transducer placed across the two
cells. As the foam in the sample cell collapses, an increase in the pressure
of this cell relative to that in the reference cell occurs as a result of the
release of the excess pressure within the foam bubbles to the headspace as
the bubbles rupture. This method of foam analysis was devised by Ross and
134 Callaghan

PRESSURE GAUGE INJECTION VESSEL

DIFFERENTIAL PRESSURE
TRANSDUCER
HYDROPUMP PISTON BOMB CONTAINING
LIVE CRUDE OIL AT
PIPELINE PRESSURE

TAP'B'

SAMPLE REFERENCE
CELL CELL

THERMOSTATTED WATER BATH

FIG. 9 High-pressure foam cell apparatus.

Nishioka [28]. A particular attraction of their method is that, in principle,


it allows the gas and liquid holdup fractions of the collapsing foam to be
determined as a function of time, as opposed to the foam height measure-
ment of the Bikerman method. These more fundamental foam parameters
can be related to the pressure above a foam collapsing in a closed vessel by
Ross' s equation of state for a foam [29]. This technique can also be used
to study foams formed from stabilised crude oils. In practice, the method
is difficult to apply as strict control of both background pressure and tem-
perature are necessary if one is to achieve reliable results. To overcome these
difficulties, an endoscope system can be incorporated into the apparatus so
as to allow visual observation of the foam under test. In this way, the foam
type and collapse characteristics can be observed under different conditions
of temperature and pressure.
An alternative method of foam height analysis has been recently intro-
duced which enables more reliable data to be obtained using a sparging tube.
This method, due to Camp and Lawrence [30], is an adaptation of that first
proposed by Ross and Cutillas [31]. It makes use of an optical transmission
foam meter l30] (see Fig. 10) in which a conventional (but ungraduated)
gas sparging tube is placed between a uniform light source and a light-sen-
Antifoams in the Oil Industry 135

LIGHT SENSITIVE
FLUORESCENT / CELL
TUBE........_

TO MAINS

TO DVM

FROSTED GLASS FOAM


(TO DIFFUSE LIGHT) TUBE
SENSITIVE CELL)

FIG. 10 Optical transmission foam cell. (From Ref. 72).

sitive cell that has good linearity in terms of voltage output against incident
light intensity. In this method the foam, created either by the conventional
sparging technique or by shaking, collapses and allows more light to reach
the light-sensitive cell. This results in a proportionate increase in voltage
output from the photocell which is coupled via an analog-to-digital converter
to a desktop computer. The voltage-time data so obtained is then analyzed
and a foam height-time curve produced from which an average foam lifetime
can be calculated. This technique has the advantage of being easier to use
than either the conventional sparge tube or the high-pressure foam cell. It
is particularly useful for the study of rapidly collapsing foams where the
conventional Bikerman technique is both difficult to use and prone to large
errors. Its principal disadvantage is that it is limited to use with stabilized
crude oils, although efforts are currently being made to develop a high-
pressure version suitable for use with live oils.
An intermediate method of assessing crude oil foam stability has been
recently developed by the Goldschmidt Company [32]. Their method allows
foam measurements to be made on simulated live crude oils at pressures up
to about 10 bar. The apparatus is shown schematically in Fig. 11 and corn-
136 Callaghan

FLOW
RATE -VENT
CONTROLLER

lo4-l----SAMPLE VESSEL

...i---WATER JACKET

t---.:1-l...-- STABILIZED CRUDE OIL


MAGNETIC FOLLOWER

CANNISTER OF
LIOUIFIED PROPANE
AND BUTANE MAGNETIC STIRRER AND HOTPLATE

PRESSURE
GAUGE

FIG. 11 Portable foam cell. (From Ref. 72.)

prises a quartz cell coupled to a liquified hydrocarbon gas canister and a


flow control valve. In this apparatus, the simulated live crude oil is created
by mixing liquified hydrocarbon gas (butane, propane) with stabilized oil.
The pressure in the system is then slowly released via a flow control valve
until it reaches the pressure at which the liquified hydrocarbon boils; a foam
is then created in a similar manner to that occurring in practice. The nature
and collapse rate of the foam in the absence and presence of antifoams can
be observed through the walls of the quartz vessel. The anti-foams are in-
troduced via a purpose-built injection cap [33].
Humphries and co-workers [34] have patented an alternative method of
determining the foaming characteristics of crude oils under simulated pro-
cess conditions. Their method uses reconstituted live crude oil obtained by
saturating stabilized oil with gas under high pressure and comprises passing
this gas-saturated oil into a flash separator. The amount of foam formed is
then determined with respect to the amount of gas withdrawn from the sep-
arator and the amount of oil contained therein. A full description of this
technique is given in the patent specification [34].

3. Methods Applicable to Fuel and Lubricating Oils


The standard method for assessing the foaming characteristics of lubricating
oils is the ASTM 0892-IP 146 test. In this test, an apparatus of the type
Antifoams in the Oil Industry 137

I
to volume meter air flow 89 to 99
ml/min

1---1-

- f-

--
:
y--
• () ..
I"'--
1
flowmeter

gas diffuser
1 L graduated cylinder stone

24'c Bath 93.5· C: Bath

FIG. 12 Apparatus for ASTM D892-IP 146 foam test.

shown in Fig. 12 is employed. It comprises a graduated glass cylinder and


an air inlet tube to the bottom of which is fastened a spherical, fused crys-
talline alumina diffuser stone of specified maximum pore diameter (>80
µm) and air permeability (3 to 6 L/min at 250 mm water pressure). Air is
blown through the inlet tube into the oil under test for a period of 5 min,
the oil being maintained at 24°C. At the end of this period, the air supply
is shut off and the foam volume recorded. The sample is then allowed to
stand for 10 min and the foam volume again recorded. The test is then
repeated on a second sample at 93.5°C, and then, after collapsing the foam,
at 24°C. The foam volume measured after a specified time is reported as
the foaming tendency (at end of 5 min blowing period) or foam stability (at
end of 10 min settling period) in milliliters. However, this test has been
shown by Watkins [35) to be intrinsically prone to a lack of reproducibility.
To overcome this problem, Watkins has devised an improved foam test [35)
based on the use of a conical, as opposed to a cylindrical, foam vessel. His
138 Callaghan

results showed that the conical geometry gave good repeatability. Recently,
Ross and Suzin have reexamined Watkins' hypothesis and concluded that
his results are sound. Furthermore, they have highlighted two points of im-
mense practical importance; namely, that the shape of the vessel used has
a positive influence on foam flooding and that the propensity toward flood-
ing is high for foam formed in a cylindrical vessel. Thus, it would seem
that cylindrical vessels are a poor choice for foaming tests if repeatability
of the test is a key factor.
4. Other Test Methods
In addition to the methods already discussed, numerous other attempts [37-
471 have been made to devise a reliable method of assessing foaming char-
acteristics. While each of these alternative methods has its own specific ad-
vantages, no single method has yet been found which can be regarded as
ideal for all situations. This is perhaps not too surprising since, according
to Bikerman [25], "an ideal measurement of the foaminess would result in
a number independent of the apparatus and the procedure being employed
and being characteristic for the solution tested (at a given temperature, etc.),
as say, viscosity or surface tension." However, if Bikerman' s ideal could
be realized then the problem of characterizing and comparing foaming sys-
tems would be greatly simplified.

IV. THE RELATIONSHIP BETWEEN PHYSICAL


PROPERTIES AND FOAM STABILITY
The stability of nonaqueous foams has been shown by numerous authors
[2-5,8,26] to be related to the following physicochemical properties:
1. the bulk rheological properties of the foaming liquid
2. the nature of the filler gas
3. the composition of the foaming system
4. the shear and dilatational properties of the gas-liquid interface
Of these four factors, the last three have been found to play the most sig-
nificant roles in antifoam selection.

A. Bulk Rheological Properties


The drainage of liquid from a single foam film can be likened to that of
liquid from between parallel plates. Using this analogy, it can readily be
seen that the approach toward a critical film thickness will be governed by
the bulk viscosity of the foaming liquid. Obviously, highly viscous liquids
will take longer to drain down to a given thickness than less viscous liquids.
Consequently, highly viscous liquids will give rise to the longer-lived foams,
Antifoams in the Oil Industry 139

all other factors being equal. Thus, it is perhaps not too surprising that for
stabilized crude oil foams, it has been found [26] that the average foam
lifetime LF is linearly related to the crude oil bulk viscosity in the following
manner:

LF = GT) 8 +b (5)
where
LF = the average foam lifetime in s
TJa = the bulk viscosity of the crude oil in mPa · s

and a,b are the constants characteristic of the filler gas used to create the
foam; e.g., for natural gas, a = 1.8 and b = 7.7.
Furthermore, it has been found that this relationship holds for all tem-
peratures up to at least 65° C [48]. Since the bulk viscosity of crude
oils (and indeed most other liquids) decreases with an increase in temper-
ature, it is expected that the foam stability of such oils should follow
a similar pattern. This is, in fact, found to be the case for stabilized crude
oils. However, for live crude oils, the situation is more complex since
the volume of gas evolved as the live oil is depressurized increases
with increasing temperature. This enhanced evolution of gas causes the
production of not only a larger quantity of foam but also, in many in-
stances, of a different type of foam to that produced at lower temperatures.
Hence, in assessing crude oil foam stability one needs to ensure that both
stabilized and live oil foam tests are carried out at the relevant operating
conditions.

B. Nature of the Filler Gas


Although bulk viscosity is a key factor in determining foam lifetime, it alone
does not control foam collapse rate. Lawrence and co-workers !49] have
investigated the mechanism of foam collapse and shown the relative im-
portance of both liquid drainage and interbubble gas diffusion. Their work
clearly shows that interbubble gas diffusion plays an important part in de-
termining foam collapse rates and that the nature of the filler gas is of prime
importance in this process. Sharovarnikov and co-workers [50] have also
investigated the role of filler gas type in determining foam stability. These
authors [50] found that for aqueous foams the stability decreased with in-
creasing filler gas solubility in the foaming liquid. Callaghan and Neustadter
[26] found the reverse effect for crude oil foams, in that stability increased
on going from air as the filler gas to natural gas. This latter finding seems
to be counterintuitive since the relative solubilities of air and natural gas in
crude oil suggest that air-filled foams should have the greater stability. It
140 Callaghan

has been suggested [26] that this difference may be due to wax particles
being produced more readily at the gas-liquid interface in the case of natural
gas foams, leading to enhanced stabilization. Attempts to test this hypothesis
have proved to be inconclusive: undoubtedly this anomaly needs to be in-
vestigated in more detail.

C. Composition of the Foaming System


The composition of a crude oil has been shown [8] to have a marked influ-
ence on its foamability. Callaghan and co-workers [8] in a series of exper-
iments with 16 different crude oils showed that the major foam-stabilizing
agents were heavy metal salts of short-chain carboxylic acids and alkyl phen-
ols of molecular weight <400. Furthermore, they showed that this same
suite of chemicals was responsible for the foaming characteristics of a wide
range of crude oils from many different sources. In many instances, this
fraction of the crude oil responsible for its foam behavior represented no
more than 0.02% w /w of the whole crude.

D. lnterfacial Rheological Properties


The rheological properties of interfacial films plays a major role in the sta-
bilization of foams. Prins and van den Tempel [51] have shown that a stable
foam possesses both a high surface dilatational viscosity and elasticity. This
has also been shown to be true for a limited number of stable crude oil foams
[15,27]. Furthermore, it has been shown [15] that the dilatational properties
of the crude oil-gas interface can be used to distinguish between profoaming
and antifoaming chemicals. The surface rheological relaxation spectra de-
rived from such interfacial dilatational measurements are useful in finger-
printing antifoam chemicals, thereby enabling the profoaming components
of an antifoam composition to be identified.
1. Techniques Used to Measure lnterfacial Rheological
Properties
a. Interfacial Shear Rheology Measurements. Numerous techniques have
been suggested for the measurement of shear rheological properties of gas-
liquid interfaces (see, for example, Joly [52]). In the case of nonaqueous
systems, the technique described by Mannheimer [11] has been shown to
be useful [26]. Basically, Mannheimer' s rheometer consists of an annular
canal with a moving floor. The surface shear viscosity is determined by
comparing the flow times of a tracer particle (e.g., talc or PTFE) placed on
the surface of a pure liquid at the centerline of the canal (see Fig. 13) to
that of a similar particle placed on the surface of a crude oil in a similar
Antifoams in the Oil Industry 141

rigid support for canal

/
trough

canal

synchronous motor

FIG. 13 Canal viscometer.

experiment. The glass trough containing the liquid (reference or crude oil)
is rotated at a constant speed, and the flow times determined for complete
revolutions of the particle in each case. The dimensions of the canal are
chosen so that the simplified equation of Burton and Mannheimer [13] can
be used, namely

TJs = YoTJB
1T
(-t_1 -)
t( - 1
(6)
142 Callaghan

where
Y0 = canal width in cm
'lls = bulk viscosity of the oil in mPa·s
t1 = time in seconds for the tracer particle to complete one revolution on
crude oil
r; = time taken for the same tracer particle to complete one revolution on
a pure liquid with the same bulk viscosity as the crude oil
'lls = surface shear viscosity of the crude oil in g/s
Using this method, it has been shown [26) that crude oil foam stability in-
creases with increasing surface shear viscosity.
Alternative methods of measuring surface shear viscosities have been pro-
posed by Davies [53], Oh and Slattery [54], and de Feijter and Benjamins
[55].
b. lnterfacial Dilatational Measurements. The dilatational modulus, E*
is defined as the surface tension increment (dy) per unit fractional surface
area change (dA/A), namely,
dy
E* =A- (7)
dA
This parameter is of both fundamental and applied interest in foam/colloid
science. Normally, data on this parameter are obtained from the properties
of longitudinal surface waves or, more directly, from the periodic surface
tension variation resulting from an imposed areal variation. Since c* is a
complex function of frequency, a scanning of the system at a number of
frequencies is usually involved in its determination [56]. However, Loglio
and co-workers [57) have developed a single pulsed Fourier transform (FT)
technique to obtain c:* as a function of frequency. In Loglio' s method, mea-
surements are made on a Langmuir trough (see Fig. 14) by expanding (or
contracting) rapidly the interfacial area by a known amount and recording
the recovery of the surface pressure with time. The resulting surface pres-
sure/time curve is then analysed using the FT method to give both the sur-
face dilatational elasticity and viscosity over a range of frequencies. An al-
ternative method of obtaining such dilatational data is the pulsed-bubble (drop)
technique as used by Lunkenheimer and co-workers [58) and Clint and co-
workers [59).
Callaghan and co-workers [ 16,26] have used the Langmuir technique of
Loglio to obtain surface dilatational data on both crude oil and crude oil/
antifoam systems. Using such data, these workers were able to differentiate
between profoamers and anti foams, and were also able to optimize anti foam
dose rates. They found that profoamers increased the dilatational elasticity,
whereas anti foams decreased it marked! y. Furthermore, they found that the
Antifoams in the Oil Industry 143

surface dilatational relaxation spectrum of a crude oil/antifoam system was


a sensitive tool in dose optimization and could be used to define profoaming
limits.

V. ANTIFOAM TYPES AND METHODS OF


SELECTION
A. Antifoam Types
The principal antifoam chemicals used in the treatment of nonaqueous foams
in the oil industry (see Pape [60]) are polydimethylsiloxanes (PDMS),
fluorosilicones, and silicone glycols. The basic structures of these materials
are given in Fig. 15.
The polydimethylsiloxane antifoams are available as 100% active odor-
less fluids; as dispersions in a range of nonaqueous carrier solvents; as com-
pounded materials consisting of silica dispersed in PDMS fluid; and as aqueous
emulsions. Of these various types of PDMS antifoam, the 100% fluids and
dispersions are the preferred forms for oil industry use, since they have low
solubility in hydrocarbon liquids and are effective at low concentrations. In
general terms, the 12,500 and 60,000 mm 2 /s grades of PDMS fluid are most
useful in oil production and refining operations, whereas the lower-viscosity
(ea. 100 mm 2 /s) fluids have some utility in oil products. The compounded
materials do not appear to be as useful in crude oil operations as they are
in aqueous systems. Indeed, it is questionable whether or not the presence
of particles is necessary for effective nonaqueous foam control. This latter
point may be worthy of further investigation. For further information on the
PDMS antifoams the reader is referred to the patent literature (see, for ex-
ample, Refs. 61-64).
The fluorosilicone materials are also available as 100% fluids and dis-
persions. These products are useful in aggressive environments, such as sour
crude treatment systems, where their greater resistance to chemical attack
and solubilization makes them highly effective. They are also useful in sit-
uations where low dosage rates are desirable. In use, these materials appear
to be less sensitive to the proportion of light end fractions in crude oil than
the PDMS antifoams. Examples of this class of antifoam can be found in
Refs. 65-68.
Silicone glycol antifoams [691 tend to be provided in 100% fluid
although they are also available as dispersions in carrier solvents. This class
of antifoams is proving to have utility in the control of diesel fuel foaming
[70].
Another class of antifoam used in the oil industry are the alkyl poly-
acrylates [71]. These materials are usually supplied in a carrier solvent such
144 Callaghan

t
TO MICROBALANCE
FIXED MOVEABLE
BARRIER BARRIER

LIQUID

LANGMUIR RAILS WILHELMY GAS- LIQUID INTERFACE


TROUGH PLATE CONTAINED BETWEEN TWO
RAILS AND TWO BARRIERS
(a)

0 0

(b)

FIG. 14 Langmuir trough for dilatational rheology measurements: (a) viewed from
the side; (b) viewed from above.
Antifoams In the 011 Industry 145

POLYDI MBTHYLSILOXANl!S

?"3
(CH3 ~ Si-O(-,IO-)nSl(CH3)3
CH2CH2CF3

FLUOROSILICONl!S

SILICONE GLYCOLS

FIG. 15 Antifoam structures.

as a petroleum distillate. Their particular area of use is in oil products such


as lubricating oils where their good air-release characteristics make them
more attractive than PDMS materials. It is possible that the better air release
observed with these materials is a direct consequence of their poorer anti-
foam performance when compared to silicone oils. The polyacrylates have
a much less pronounced effect on surface elasticity than the PDMS materials
and, as a result, are less likely to reduce the beneficial circulatory flows in
the rising bubble surface which aid air release.
146 Callaghan

TABLE 1 Antifoams and Diluents Commonly Used in Oil Industry Processes

Antifoam type Application(s) Diluent(s)

Polydimethylsilicone Gas-oil separation White spirit


(1,000, 12,500, or Vacuum distillation Kerosene
60,000 mm 2 /s Delayed coking Diesel oil
grades) Crude distillation Toluene
Asphalt extraction Xylene
Visbreaking MEK
pet. ether
Polydimethylsilicone Lube oils As above
(< 1,000 mm 2 /s
grades)
Compound silicone (sil- Propane deasphalting Toluene
ica dispersed in sili- Naphtha
cone oil) Amyl acetate
Fluorosilicone Diesel fuel 2-Ethoxyethyl acetate
Sour crude processing Acetone
MEK
White spirit
Silicone glycol Diesel fuel Toluene
Gas oil
Alkyl polyacrylate Lube oil Petroleum distillate

Table I gives an outline of the antifoam and diluent types used for a
number of oil industry applications. The influence of added solvents on an-
tifoam performance has been discussed by Callaghan and co-workers [72].

B. Selection of Antifoams
In order to be an effective antifoam, a material must meet four basic
requirements:
l. It must be insoluble in the foaming medium under the relevant process
conditions.
2. It must have a lower surface tension than the foaming medium.
3. It must be rapidly dispersed in the foaming medium.
4. It must be chemically inert.
Insolubility of the antifoam compound in the foaming medium is very im-
portant, since an antifoam with its low surface tension will actually become
a profoamer if soluble in the foaming medium. Silicones with their low
surface tension and low interfacial tension against hydrocarbon oils, are ide-
ally suited to enter and spread in nonaqueous systems and, as such, are ideal
Antifoams in the Oil Industry 147

antifoams. However, the selection of the precise silicone required needs to


be carried out carefully so as to avoid any possible profoaming problems.
The preferred method for selecting an antifoam for a specific process is
as follows:
a. Determine the foam type in a process simulator.
b. Measure the surface tension of the foaming liquid, preferably under
simulated operating conditions.
c. Rapidly screen a number of potential antifoam candidates, using surface
tension as an initial sifting tool, using the Bikerman sparge tube de-
scribed earlier.
d. Determine the surface rheological spectra of the best candidates selected
in (c).
e. Optimize dosage, and define profoaming limits (if any), using both sur-
face rheological and Bikerman foamability measurements.
f. Finally, check the performance and profoaming limit of the chosen an-
tifoam in a process simulator prior to recommendation.
The effect of both profoamers and antifoams on the surface rheological re-
laxation spectra of crude oil-gas interfaces has been described in Ref. 16.

VI. FUTURE DIRECTIONS


In general terms, the antifoams currently available can tackle most oil in-
dustry problems satisfactorily. However, two areas where improvements could
still be made are sour crude oil defoaming and inhibiting foam formation in
oil products, especially diesel fuel and lubricating oils. In the latter case,
the problem is to produce an antifoam that is effective at low ( <5 ppm)
concentration and yet does not adversely affect the oil's air-release
characteristics.

ACKNOWLEDGMENTS
I would like to thank The British Petroleum Co. plc for permission to publish
this chapter, and my colleagues, Messrs. B. J. Oswald and G. J. J. Jayne,
for their counsel during its preparation.

REFERENCES
I. S. Ross, Chem. Eng. Prog. 63: 41-47 (1967).
2. A. P. Brady and S. Ross, J. Am. Chem. Soc. 66: 1348-1356 (1944).
3. J. W. McBain and J. V. Robinson, Natl. Advisory Comm. Aeronaut., Tech.
Note No. 1844, 1949.
148 Callaghan

4. J. W. McBain, S. Ross, A. P. Brady, J. V. Robinson, I. M. Abrams, R. C.


Thorburn, and C. G. Lindquist, Natl. Advisory Comm. Aeronaut., War time
Report, ARR no. 4105, 1944.
5. S. Ross and R. M. Haak, J. Phys. Chem., 62: 1260-1264 (1958).
6. R. J. Mannheimer, and R. S. Schecter, J. Colloid Interface Sci. 32: 212-224
(1970).
7. A. Scheludko and E. Manev, Trans. Faraday Soc. 64: 1123-1134 (1968).
8. I. C. Callaghan, A. L. McKechnie, J. E. Ray, and J. C. Wainwright, Soc.
Petrol. Eng. J., 171-175 (1985).
9. S. Ross and J. W. McBain, Ind. Eng. Chem. 36: 570-573 (1944).
10. I. B. Ivanov, B. Radoev, E. Manev, and A. Scheludko, Trans. Faraday Soc.
66: 1262-1273 (1970).
11. R. J. Mannheimer, A. I. Ch. Eng. J. 15: 88-93 (1969).
12. J. W. McBain, J. V. Robinson, W. W. Woods, and I. M. Abrams, Natl.
Advisory Comm. Aeronaut., Tech. Note No. 1845, 1949.
13. R. A. Burton and R. J. Mannheimer, Adv. Chem. Ser. 63: 315-328 (1967).
14. J. Lucassen and M. van den Tempel, Chem. Eng. Sci. 27: 1283-1291 (1972).
15. G. Loglio, U. Tesei, and R. Cini, J. Colloid Interface Sci. 71: 316-320 (1979).
16. I. C. Callaghan, C. M. Gould, R. J. Hamilton, and E. L. Neustadter, Colloids
Surf. 8: 17-28 (1983).
17. G. Loglio, U. Tesei, and R. Cini, J. Colloid Interface Sci. JOO: 393-396
(1984).
18. J. V. Robinson, and W. W. Woods, J. Soc. Chem. Ind. 67: 361-365 (1948).
19. S. Ross, J. Phys. Colloid Chem. 54: 429-436 (1950).
20. W. D. Harkins and E. Boyd, J. Phys. Chem. 45: 20-43 (1941).
21. L. T. Shearer and W. W. Akers, J. Phys. Chem. 62: 1269-1270 (1958).
22. A. D. Barber and E. F. Wijn, Proc. 3rd International Symposium on Distil-
lation, 2: 3.1/15-3.1/35, 1979.
23. E. Manegold, Schaum, strassenbau, chemie und technik, 83, Heidelberg (1953).
24. I. C. Callaghan, C. M. Gould, A. Reid, and D. H. Seaton, J. Petrol. Tech.
37: 2211-2218 (1985).
25. J. J. Bikerman, Foams, Springer-Verlag, New York, 1973.
26. I. C. Callaghan and E. L. Neustadter, Chem. Ind. 53-57 (1981).
27. I. C. Callaghan and C. M. Gould (assigned to The British Petroleum Co. pie),
GB 2149127A; June 5, 1985; filed October 29, 1984.
28. S. Ross and G. M. Nishioka, J. Colloid Interface Sci. 81: 1-7 (1981).
29. S. Ross Am. J. Phys. 46: 513-516 (1978).
30. M.A. Camp and F. T. Lawrence (assigned to The British Petroleum Co. pie),
GB 2158574A; November 13, 1985; filed May 2, 1985.
31. S. Ross and M. J. Cutillas, J. Phys. Chem. 59: 863-866 (1955).
32. H. Lattek, H.-F. Fink, and G. Koerner (assigned to Th. Goldschmidt AG),
GB 2154738A; September 11, 1985; filed December 14, 1985.
33. F. T. Lawrence (assigned to The British Petroleum Co. pie), GB 2176602A;
December 31, 1986; filed June 12, 1986.
34. C. L. Humphries, E. F. Schultz, and A. M. Winkelman (assigned to Mobil
Oil Corporation), US 4426879; January 24, 1984; June 21, 1982.
Antifoams in the Oil Industry 149

35. R. C. Watkins, J. Inst. Petrol. 59: 106-113 (1973).


36. S. Ross and Y. Suzin, Langmuir 1: 145-149 (1985).
37. E. G. King, J. Phys. Chem. 48: 141-154 (1944).
38. G. L. Clark and S. Ross, Ind. Eng. Chem. 32: 1594-1598 (1940).
39. M.A. Amerine and L. P. Martin, Ind. Eng. Chem. 34: 152-157 (1942).
40. P. P. Gray and I. Stone, Wallerstein Labs. Commun. 3: 159-171 (1940).
41. E. J. Burcik, J. Colloid Interface Sci. 5: 421-436 (1950).
42. W. B. Hardy, Proc. Roy. Soc. (London) 86A: 610-635 (1912).
43. W. E. Glausser, Chem. Eng. Prog. 60: 67-68 (1964).
44. S. Hartland, and A. D. Barber, Trans. Inst. Chem. Eng. 52: 43-52 (1974).
45. H. Tsuge, J. Ushida, and S. Hibino, J. Colloid Interface Sci. 100: 175-184
(1984).
46. Deutsche Industrie Norm (DIN) Test no. 53902, Teil 1, 1981.
47. S. Okazaki and S. Sasaki; Tenside 3: 115-118 (1966).
48. I. C. Callaghan and F. T. Lawrence, unpublished results.
49. I. C. Callaghan, F. T. Lawrence, and P. M. Melton, Colloid Polym. Sci. 264:
423-434 (1986).
50. A. F. Sharovamikov, V. N. Tsap, A. Ya. Korol'chenko, and A. V. Ivanov,
Kolloid. Zh. 43: 808-812 (1981).
51. A. Prins and M. van den Tempel, Proc. IVth Int. Congress Surface Active
Subst., 2: 1119-1131, 1964.
52. M. Joly, Surface Colloid Sci. 5: 1-77 (1972).
53. J. T. Davies, Proc. Int. Congress Surface Activity, 2nd., London, I: 220-
224 (1957).
54. S.-G. Oh and J.C. Slattery, J. Colloid Interface Sci. 67: 516-525 (1978).
55. J. A. de Feijter and J. Benjamins, J. Colloid Interface Sci. 70: 375-382 (1979).
56. J. Lucassen and M. van den Tempel, Chem. Eng. Sci. 27: 1283-1291 (1972).
57. G. Loglio, U. Tesei, and R. Cini, Ber. Bunsenges. Phys. Chem. 81: 1154-
1156 (1977).
58. K. Lunkenheimer, K.-D. Wantke, and R. Miller, ABH Akad. Wiss. DDR,
445-453 (1976).
59. J. H. Clint, E. L. Neustadter, and T. J. Jones, Dev. Petrol. Sci. 13: 135-148
(1981).
60. P. G. Pape, J. Petrol. Tech 35: 1197-1204 (1983).
61. I. C. Callaghan, H. F. Fink, C. M. Gould, G. Koerner, H-J. Patzke, and C.
Weitmeyer (assigned to The British Petroleum Co. pie and Th. Goldschmidt
AG), US 4564665; January 14, 1986; filed April 5, 1983.
62. R. E. Moeller (assigned to General Electric Co.), GB 1523654A; September
6, 1978; filed October 21, 1975.
63. A. Guillaume and F. Sagi (assigned to Rhone-Poulenc Industries), GB 15 I 9541A;
August 2, 1978; filed June 6, 1975.
64. K. W. Farminer (assigned to Dow Coming Corp.), FR 2379307; September
1, 1978; filed July 4, 1977.
65. J. W. Keil (assigned to Dow Coming Corp.), US 4537677; August 27, 1985;
filed November 5, 1984.
150 Callaghan

66. H. F. Fink, G. Koerner, R. Berger, and C. Weitmeyer (assigned to Th. Gold-


schmidt AG), DE 3635093C; March 10, 1988; filed October 15, 1986.
67. D. Boemer, H. F. Fink, G. Koerner, and G. Rossmy (assigned to Th. Gold-
schmidt AG), DE 2444073B; August 19, 1976; filed March 25, 1976.
68. E. R. Evans (assigned to General Electric Co.), US 4329528; May 11, 1982;
filed December l, 1980.
69. I. C. Callaghan, C. M. Gould, and W. Grabowski (assigned to The British
Petroleum Co. pie), US 4711714; December 8, 1987; June 21, 1985.
70. Dow Coming Corp., US 4690688; September 1, 1987; filed July 7, 1986.
71. J.E. Fields, (assigned to the Monsanto Co.), US 3166508; January 19, 1965;
filed January 16, 1963.
72. I. C. Callaghan, S. A. Hickman, F. T. Lawrence, and P. M. Melton, RSC
Special Publication No. 59, 48-57 (1986).
3
Defoaming in the Pulp and Paper
Industry

S. LEE ALLEN McGill University, Montreal, Quebec, Canada


LAWRENCE H. ALLEN Pulp and Paper Research Institute of Canada,
Pointe Claire, Quebec, Canada
TED H. FLAHERTY Dorset Industrial Chemicals, Ltd., Chateauguay,
Quebec, Canada

I. Overview of the Industry 151


II. Sources of Foam 155
III. Problems Caused by Foam 159
IV. Physical Control of Foam 161
v. Chemical Control of Foam 164
A. Early antifoams 164
B. Development of brownstock washer antifoams 164
C. Other pulp and paper antifoams 168
D. Disadvantages of chemical defoaming 169
VI. Measurement of Foam 170
VII. Testing of Antifoams 171
VIII. Handling and Use of Antifoams 171
IX. Future Research and Development 173
References 173

I. OVERVIEW OF THE INDUSTRY


The pulp and paper industry is the world's biggest single user of defoaming
agents. Forecasts of pulp and paper capacities [1] and typical antifoam use
rates for various processes [2] suggest that a conservative estimate of world-
wide antifoam consumption in 1990 is 160,000 tonnes. Indeed, antifoams
are the largest single class of process control chemicals in the industry [2].
Theoretically, the production of paper from wood is simple. Wood fibers
are dispersed chemically or mechanically and suspended in water to form a
151
152 Allen et al.

Chips
In

'
Knotters Rotary Washers Screens

Digester Blow Tank

Centrifugal Deckers l,___ _ _ _ _ _ _ __


Cleaners (Thickeners) v~---------'
Multi-Stage Bleachery To
Pulp Machine
or
Paper Mm

FIG. 1 Flowchart of a typical bleached kraft pulp mill. Only pulp flow is indicated.

pulp slurry. The slurry is deposited on a fine screen, creating a fiber web
which is drained and dried, resulting in paper. In practice, the process is
considerably more complicated. The manufacture of a single type of paper,
in a specific pulp and paper mill, may require dozens of chemicals for pulp-
ing, bleaching, process control, and modification of end properties; each
different grade of paper will require a different chemical recipe and process
conditions.
In chemical pulp production, wood chips are cooked at elevated temper-
atures in solutions of various chemicals in pressurized vessels called diges-
ters. In sulfite pulping the cooking solution is composed of sulfur dioxide
and a metal ion hydroxide (Ca2+, Mg 2 +, Na+; also NH:); depending on the
process, the conditions range from pH l and 120°C to pH 9 and 180°C. The
kraft process uses sodium hydroxide and sodium sulfide solutions, at 180°C
and pH above 13. A typical kraft pulp mill is outlined in Fig. 1.
In both processes, the chemicals soften and degrade the lignin (Fig. 2)
(the material that binds the fibers together in the wood), liberating the in-
dividual fibers. After pulping, the cooked wood chips are usually "blown"
from the digester under pressure into a blowtank. From there, the pulp pro-
ceeds to the knotters, which remove the largest pieces of undispersed ma-
terial. The pulp then typically goes to a series of rotating drum washers,
where it forms a thick mat on the screen which forms the surface of each
Defoaming in the Pulp and Paper Industry 153

c?.;
9K.o!OH 9K.o!OH
CH
~H

¥0CH3 CH,c:J'.c$locH,
OH OH
Coniferyl Alcohol Sinapyl Alcohol
-Softwoods -Hardwoods
LIGNIN MONOMERS

9K.o!OH
[Llgnin]-0-CH

~=~:ignln]
0
[Lignin]/

FORMATION OF CROSS-LINKED
LIGNIN POLYMER
FIG. 2 Representative structural formulae of lignin monomers and lignin.

drum. The spent cooking liquor (black liquor) is drained off (usually by
suction applied from inside the washer drum), as the pulp mat is sprayed
with recycled process water.
It is here in the washing of unbleached pulp that over 60% of the de-
foaming agents are used [2]. To make the kraft process economically viable,
spent pulping chemicals must be recovered as efficiently as possible; hence,
thorough washing is essential. In addition, in pulp that will be bleached,
good washing is necessary to remove undesirable components that would
otherwise consume bleaching chemicals. Foam hampers the washing pro-
cess, and must be controlled to ensure efficient washing and good chemical
recovery. Ironically, it is usually during the washing of the kraft browns tock
(freshly cooked pulp) that the worst foam problems are encountered. An-
tifoam may be added immediately before the washers, at the knotters, or at
any other appropriate point in the washing sequence. Antifoam can also be
154 Allen et al.

Machine Fan
Pulper Beaters/ Screens Chest Pump
Refiners

Wire Pit

FIG. 3 Flow chart of a typical fine paper mill. Only pulp/paper flow is indicated.

sprayed across the pulp mat on the washers. A typical use rate is about l
kg/t dry pulp.
In the screen room, screens permit individual fibers to pass through, but
fiber bundles (shives) are retained. Antifoam is sometimes required. Extra
make up water is added at the screens, and the screen room filtrate is cus-
tomarily recycled back to the washers. The pulp then goes to the cleaners,
where small dirt specks are centrifugally separated from the fibers.
If a white pulp is desired, the pulp proceeds to the bleach plant, where
residual lignin and color are removed by sequential treatments with chem-
icals such as oxygen, chlorine, sodium hydroxide, hydrogen peroxide, chlo-
rine dioxide, and sodium hypochlorite. After each stage of bleaching, the
pulp is usually washed to remove residual chemicals. Foam problems may
be encountered here too.
The bleached pulp may be further screened and cleaned, and sent directly
to the paper mill; alternatively, it may go to a pulp-drying machine, if it is
to be sold as pulp. On the pulp machine the pulp slurry is distributed con-
tinuously onto a mesh, either a horizontal mesh belt or a drum resembling
a washer, where it forms a thick fiber mat. The pulp mat is then dewatered,
dried, and baled for shipment or storage.
In the paper mill (Fig. 3), the pulp may be beaten or refined to increase
strength in the final product. Then additives are mixed into the pulp slurry.
Defoaming in the Pulp and Paper Industry 155

These vary with the type of paper being made; a list of the more common
ones is given in Sec. II. Such additives frequently cause, or stabilize, foam;
also, air may be entrained when they are mixed into the pulp suspension.
On the paper machine, the pulp enters the headbox, where it is agitated
for optimum dispersion. The pulp is dispensed through a slit ("slice") in the
headbox onto a moving screen ("wire") on which the web is formed and
partly dewatered. The wet web is further dewatered in the nips of the press
section, and is dried on the steam-heated cylinders of the dryer section. The
paper may then be calendered (to increase surface smoothness) and/or coated.
Foam may be generated in any of the wet stages of papermaking and coat-
ing, and must be controlled to maintain uniform quality in the finished product.
In contrast to chemical pulping, mechanical pulping separates the wood
into its component fibers by pressing logs against rotating grindstones, or
by refining wood chips with disc refiners (which damages the fibers less
than grinding). The lignin is not dissolved, remaining largely with the fibers.
Mechanical pulps are not usually washed, but are screened and cleaned.
Although foam problems can be severe, they are generally less than with
chemical pulps.
To reduce fiber damage during mechanical pulping, the lignin in the wood
may be softened before refining. In thermomechanical pulping, the chips
are heated with pressurized steam, then immediately sent hot to the refiners.
In chemimechanical pulping, a partial chemical cooking precedes refining.

11. SOURCES OF FOAM


The three materials necessary for foam production are a liquid, a gas, and
a surfactant. In the pulp and paper industry, the liquid is water, the gas is
air, and the surfactants are numerous.
The major source of surfactants in the pulp mill is the wood itself. A
typical softwood supplied to a mill might consist of approximately 45-50%
fiber, 20-30% lignin, 20-35% miscellaneous carbohydrates, and 1-8% wood
resin, based on oven-dried weight [3,4). The worst foaming generally occurs
after chemical pulping, where the spent pulping liquors must be washed
from the fibers. Table 1 gives examples of compositions of spent pulping
liquors; the surface active materials are found predominantly in the lignin
and wood resin fractions.
Chemical additives also add to the surfactant load, chiefly in the paper
mill. Table 2 summarizes the main classes of additives and the types of
chemicals used in each class. No single papermaking furnish (the mixture
of pulp and chemicals supplied to the paper machine) will contain all of
these additives; for example, wet strength resins are used only for paper that
must hold up under damp or humid conditions, such as tissue and towel.
156 Allen et al.

TABLE 1 Compositions of Spent Pulping Liquors

Type of %
liquor• Class of material Composition Solidsb

Kraft black Lignin Monomeric and polymeric 33


liquor lignin with a broad range of
12-15% molecular weights; some sub-
solids stitution; lower molecular
weight fraction dissolved
Organic acids >90% monocarboxylic acids 27
Most polyhydroxy acids
Inorganic Spent pulping chemicals, 23
components plus small amounts from
wood: NaOH, Na 2S,
NaHS04, Na2S04,
Na2C0 3 some K+,
Ca2+, Cl-
Extractives Wood resin (60-90% saponi- 6
fied): fatty and resin acids,
di- and triglycerides, steryl
esters, sterols, flavonoids,
etc.
Bound sodium Sodium bound to lignin, or- 11
ganic acids, and extractives
Spent sulfite Lignin Monomeric and polymeric 44
liquor lignosulfonates with a broad
11-14% solids range of molecular weights
Carbohydrates >80% monosaccharides; 23
mainly xylose, mannose, ga-
lactose, glucose
Inorganic Spent pulping chemicals, 16
components plus small amounts from
wood (usually sodium salts):
NaHS03, Na2S03, Na2S04,
Na2 S20 3 , Na 2C0 3 depending
on process, Mg 2 + or Ca2+
salts K+, Cl-
Organic acids Acetic and aldonic acids 7
Extractives Wood resin (not saponified) 7
Bound sodium (ea!- Sodium (calcium, magne- 3
cium, magnesium) sium) bound to
lignosulfonates
'Material exiting from digester = 7-10% fiber, balance spent liquor.
bPercent composition for specific examples only. Actual percentages will vary greatly de-
pending on the wood and the pulping conditions.
Source: Ref. 5.
Defoaming in the Pulp and Paper Industry 157

TABLE 2 Paper Mill Additives

Class Typical chemicals used

Fillers Ti0 2
Clay (aluminosilicate platelets with nega-
tive faces and positive edges; usually
kaolin)
Retention aids Polyacrylamide
Polyethyleneimine
Pol ye thy leneoxide
Polyacrylic acid
Wet strength resins Cationic resins
urea formaldehyde
melamine formaldehyde
polyamideamine epichlorhydrin
Starch Polymers of glucose
(dry strength) amylose (linear)
amylopectin (branched)
may be anionic or cationic
Internal size Rosin size (mixture of resin acids)
(water repellency Fortified rosin size (maleic anhydride or
for good printing) fumaric acid added)
Alkaline size Alkyl ketene dimer
Alkenyl succinic anhydride
Slimicides Organic biocides
Cl 2 , Cl0 2 , NaClO
Resin deposition Alum
(pitch) control additives Sodium aluminate
Talc
Dispersants:
pleuronics (ethyleneoxide/
propylene-oxide copolymers)
alky !phenol/ethy leneoxide
condensates
aliphatic or aromatic sulfated or sulfon-
ated esters, oils, and phenolic
derivatives
polyacrylic acid
Dyes and pigments Acid dyes (require alum)
Basic dyes
Direct dyes
Pigments
dispersions of colored particles
Ti0 2 (white)
158 Allen et al.

TABLE 3 Example of Paper Mill Formulation: Coated Publication-Grade Paper


from Groundwood and Kraft Pulps

Additive Use kg/tonne of paper

Clay Filler; opacifier 225


Cationic starch Retention; strength 43
Talc Control of deposits (pitch 20
control)
Alum pH control; pitch control 5
Sodium hydroxide pH control l
Cationic polyacrylamide Retention 0.3

An example of a papermaking furnish is shown in Table 3; actual formu-


lations vary widely, and depend greatly on the type and grade of paper being
made. Many of the additives attach to the fibers and end up in the paper,
but certain quantities remain in the white water (the liquid which drains off
the fiber web on the paper machine).
Even when the pulp is largely free of additives (e.g., newsprint) the white
water contains considerable amounts of dissolved and suspended solids. Ta-
ble 4 presents analyses of three newsprint white waters, showing substantial
quantities of organic and inorganic material dispersed in the water.
Particulates in the process liquid, such as fines (fiber fragments and ray
cells that can pass through a paper machine wire) and fillers, although not
actually surface active, can stabilize foam under some conditions, or con-
tribute to the residue that dry foam leaves on equipment. Temperature and
pH can also influence foam formation and stability, but it is difficult to
generalize about their effects because of the complexity and variability of
the process liquids.
Cascades and vigorous agitation are among the most obvious ways of
introducing air into the stock (process liquid plus pulp). Rapid expulsion of
cooked pulp from pressurized digesters causes foam, both by the release of
dissolved gases when pressure is reduced, and by uptake of air in the ensuing
turbulence. The problem at this point is aggravated by the high concentra-
tions of resin and lignin dissolved in the cooking liquor. Leaking pumps,
pipe fittings, and washer seals permit air to enter the system. Addition of
cold water to warm stock and heating of cold stock release dissolved gases.
Vacuum drainage of process water through screens can also release dis-
solved air by pressure reduction, and can introduce additional air, if air is
drawn through the screen with the water. Dry material mixed directly with
the process water can carry air into the stock. The amount of air entrained
at a given point will depend on the nature of the process equipment and the
Defoaming in the Pulp and Paper Industry 159

TABLE 4 Compositions of Newsprint White Waters

Mill A Mill B Mill C

Type of pulp Thermo- Thermo- 72% Stone


mechanical mechanical groundwood
28% Chcmi-
mechanical
pH 4.8 4.6 4.3
Lignin (g/L) 0.90 0.95 0.15
Dissolved solids (g/L) 3.58 7.00 1.09
Suspended solids (mg/L) 80 198 33
Ca2 + (mg/L) 21.8 54.0 14.8
AIH (mg/L) 1.10 1.82 0.88
Mg 2 + (mg/L) 5.68 20.9 5.03
Total S (mg/L) 282 370 70
S03 - (mg/L) 144 36 nil
S0 4 2 - (mg/I) 672 954 207
Free sugars (mg/L) 79 74 nil
Hemicellulose (mg/L) 879 3852 266
(polymeric sugars from
degraded cellulose)
Colloidal wood resin 2.0 x 106 2.3 x 106 0.13 x 106
(particles/mL)
Source: Ref. 6.

manner in which it is operated, and cannot be predicted unless one is fa-


miliar with on-site conditions.

Ill. PROBLEMS CAUSED BY FOAM


The most noticeable type of foam is that floating on the surface of the stock.
In some respects, this foam is easier to handle, as the problem and its treat-
ment can be monitored readily. Less obvious, and harder to monitor, is foam
in the form of small bubbles (1-100 J.1m in diameter) entrained in the bulk
of the liquid or adhering to the solid components of the stock.
Some of the problems caused by foam are obvious. The volume of the
foam reduces the effective capacity of pumps and tanks, and thus either
reduces output, or cuts down on the residence time in washers, refiners,
mixing tanks, bleaching towers, etc. Solids, such as fibers, fines, or fillers,
or surface active residues or additives, can be concentrated in surface foam;
overflow can then result in loss of material and changes in relative propor-
tions of components, as well as a mess and a possible safety hazard (Fig.
160 Allen et al.

FIG. 4 Overflow of surface foam in a pulp mill.

4). If the level of the surface foam drops, foam left on the walls of a vessel
may leave behind deposits that can break off and contaminate the pulp.
As mentioned previously, one of the biggest foam-related problems in
chemical pulp (especially kraft) mills is maintaining rate and efficiency of
washing on rotary drum washers. As wash water is drawn through the pulp
by partial vacuum in the interior of the drum, air bubbles in the stock can
give rise to channeling. These channels reduce washing efficiency in two
ways: if air passes through the channels, the vacuum is reduced and less
black liquor is removed; if wash water from the showers is channeled through
the mat, it is less effective at displacing the black liquor from the pulp.
Physical problems are encountered when handling foaming liquids. Pumps
can cavitate when attempting to pass large volumes of foam. Metering de-
vices are rendered less accurate. Liquid held in surface foam may not be
blended thoroughly into the bulk of the stock during mixing.
In the paper mill, foam may reduce beater and refiner efficiency by cush-
ioning the mechanical action on the pulp [7]. Drainage efficiency on the
paper machine is reduced by channeling. Bubbles breaking on the fiber web
leave behind traces of solids (fibers, fines, fillers, pigments, dirt, pitch, etc.)
Defoaming in the Pulp and Paper Industry 161

FIG. 5 Foam flaws in fine paper. Small air bubbles in the stock on the paper
machine wire result in the appearance of circular thin spots in the final product. The
spots were encircled by the quality control technician who rejected the paper.

that mark the sheet or cause it to stick to rolls and rip. If the foam bubbles
displace fiber significantly thin spots result, leading to the speckled ap-
pearance of the paper shown in Fig. 5; in extreme cases, the web breaks.
Foam in paper coating media may result in heterogeneous application and
bubble spots (Fig. 6).
Surface foam in effluent is aesthetically unacceptable, and may hamper
effluent treatment by impeding settling of particles held in the foam, and by
reducing the rate of oxygen uptake by the effluent. Occasionally, foam can
be used to advantage by mill personnel; the foam with its high concentration
of surfactants (lignin, lignosulfonates, and/or resin [8,9)) can be skimmed
off and treated separately.

IV. PHYSICAL CONTROL OF FOAM


The first steps in physical prevention of foam are relatively inexpensive.
Elimination of cascades may require only simple repiping; closing up leaks
in pumps, fittings, and seals is standard maintenance. Further equipment
162 Allen et al.

FIG. 6 Scanning electron micrograph of foam flaws in paper coating. White bar
at bottom denotes 100 µm. The small white spots are intact dried bubbles in the
coating. The dark circular areas are holes in the coating caused when the tops of
larger dried bubbles are broken, probably in calendering. (Photograph courtesy of
Glynis de Sylveira.)

modifications to avoid air entrainment often involve large capital investments.


The extensive foam problems in kraft brownstock washers, and the con-
sequent costs of chemical antifoams required for foam control, have been
reduced by changes in brownstock washing equipment. Most of the new
equipment is added to the process line just upstream of the conventional
rotary drum washers. Displacement washing in the bottoms of continuous
digesters removes significant quantities of foaming materials with the black
liquor, without air entrainment. Conventional blowing of the pulp slurry out
of batch digesters is effectively a violent cascade; dissipating digester pres-
sure more gradually [ 10], or ejecting the pulp so that it does not come in
contact with air (for example, into the bottom of a vessel filled with stock)
avoid this. Diffusion washers [11], in which wash liquor is pumped through
the stock, further reduce air uptake, as do pressure knotters. These changes
in process equipment decrease the amounts of air and surfactant in the stock,
Defoaming in the Pulp and Paper Industry 163

Spray Nozzles

To
Vacuum-- ,,.,
\II•
.
,.,.,..
'. ,
... :\ /,~
"111.
I
1\i/
.. /
·.. ..,,
'.!/
' . ::;..(;. '·\\!/:' .
-.:•J/
Overflow
Weir
L_

i
Overflow
-Recirculated
to Inlet t
Outlet - to-Headbox
Supply Pump
(Fan Pump)

FIG. 7 Schematic of the Deculator vacuum deacralor (manufactured by Clark and


Vicario). This apparatus is usually installed just upstream of the paper machine.
Stock is sprayed into an evacuated chamber, entrained air boils off, and the deaer-
ated stock passes through the fan pump to the headbox.

and thus reduce the amount of foam (and the need for chemical antifoams)
in the washers. Further foam reduction has been reported when drum wash-
ers are replaced by horizontal pressure washers [ 12]. The new machinery
designs have been aimed primarily at improving washer efficiency, with
consequent improved product quality and decreased chemical losses and bleach
chemical use; foam prevention was pursued as much for improved washing
as for decreased cost of chemical anti foams.
Innovation has also reached the screen room, where pressure screens arc
reported to reduce foaming [ 13].
Commercial apparatus is available which breaks surface foam by beaters,
bars, and sonic fields; entrained foam can also be removed by centrifugal
force and vacuum deaeration. A sketch of a common vacuum deaerator (the
Deculator) is shown in Fig. 7. Physical defoaming equipment is most useful
when the foam is consistent, copious, and localized.
Water showers can be used to break surface foam. This does little to
reduce foam downstream from the spraying site, and may significantly dilute
the pulp. Nevertheless, it is sometimes the method of choice to control bub-
bles in the headbox of the paper machine.
In treating foam physically, one must consider the initial cost of equip-
ment, lack of flexibility in treatment site, and lack of carrythrough of the
164 Allen et al.

defoaming effect downstream. If the technique, the situation, and the point
of treatment are chosen carefully, mechanical foam control has the advan-
tages that it is relatively insensitive to chemical variations in the stock, and
it adds no new chemicals to an already complex system.

V. CHEMICAL CONTROL OF FOAM


A. Early Antifoams
The first chemical antifoams were aimed exclusively at breaking visible sur-
face foam. Milk and cream were the forebears of modern emulsion-type
antifoams. Kerosene, fuel oil, and other light petroleum oils enjoyed long-
term popularity for their ability to break foam; they are still used occasion-
ally, although odor, flammability, environmental pollution, and deteriora-
tion of rubber rolls and fittings are drawbacks. Pine oil and linseed oil found
some use. Alcohols with C7-C 16 chains were effective but expensive an-
tifoams, and were often mixed with kerosene for economy (and sometimes
worked better for it).
In the 1950s, oil-in-water emulsion* antifoams were developed using fatty
acids, fatty alcohols, esters, waxes, sulfated tallows, and saturated soaps
[14]. Some were sold as preemulsified liquids and pastes, thus ensuring
good control of emulsification but increasing the cost of shipping to the
customer. Others were emulsifiable liquids and solids that were cheaper to
ship but which had to be emulsified by the user, with less consistent results.
Both types were dispersed in water before use.

B. Development of Brownstock Washer Antifoams


Also during the 1950s, silicone antifoams (polydimethyl siloxane oil, dis-
persed in water or light oil) were introduced to the industry. Their use on
paper machines and in the latter stages in pulp mills was limited by the
tendency of silicone in the paper to interfere with physical properties, print-
ing, and coating [7]. The stability of these antifoams at high temperature
and alkalinity, however, made these the first truly effective antifoams for
kraft black liquor and brownstock washers (temperature up to 90°C, pH up
to 12-13).
Dispersions of hydrophobic particles in light oil formed the next gener-
ation of brownstock washer antifoams. The first such antifoams were pat-

*No distinction is made in the pulp and paper antifoam literature between true emulsions and
dispersions of a liquid crystal phase in a bulk liquid. Here, for simplicity, the term emulsion
will refer to any colloidal dispersion of one liquid in another, regardless of the actual phase
of the dispersed liquid.
Defoaming in the Pulp and Paper Industry 165

ented in 1963 [15); they consisted basically of silica particles, rendered hy-
drophobic by a high-temperature treatment with silicone, and dispersed in
mineral oil. The early 1970s saw the introduction of oil-based brownstock
washer antifoams using hydrophobic amide wax particles (e.g., [ 16]). Some
antifoams contain both hydrophobic silica and amide wax particles, in light
oil.
Just how these antifoams work is not fully understood (see Chapter 1 for
discussion of mechanisms). Most workers are in agreement that the partic-
ulate components act by entering and disrupting the foam film [ 17], but the
role of the oil is less well defined. Much of the patent literature refers to
the oil as an "inert liquid" or "carrier oil," implying that it plays no active
part in defoaming. However, the success of kerosene and other light oils as
antifoams suggests that the oil in these brownstock washer antifoams may
be a major contributor to the defoaming process.
The oil crisis of the early 1970s fueled the push to reduce the amount of
oil in antifoams. "Water-extended" antifoams appeared, which were at first
essentially oil-based antifoams with water emulsified into the oil phase, ef-
fectively diluting the concentration of solid defoaming particles. They were
less expensive on a weight basis, but less efficient. Now, water-extended
antifoams are similar to the oil-based compositions in solids content, but up
to half of the oil is replaced by water (plus appropriate stabilizers and emul-
sifiers). When more than half of the oil is replaced with water, the antifoam
becomes an oil-in-water emulsion; these formulations are termed "water
based."
Table 5 gives a representative sampling of patents for brownstock washer
antifoams [15,16,18-32), beginning with the first hydrophobic silica par-
ticle antifoams. Unless indicated otherwise, the preferred bulk liquid is a
light petroleum oil (mineral oil). In many patents the preferred formulation
contains no water, but water-extended and/or water-based formulations are
also included, giving the option of eliminating part of the oil if desired.
It is difficult to relate patents to antifoams in the marketplace. Compe-
tition makes manufacturers reluctant to give detailed information about their
formulations; when a product can be definitely associated with a specific
patent, the range of formulations covered by the patent makes it impossible
to determine exactly how the antifoam is made. Usually all that can be de-
termined is the type of antifoam (e.g., silica, wax, or silica-wax) and the
nature of the liquid base (e.g., oil-based, oil-based/water-extended, or water-
based).
This lack of precise information on the compositions of commercial an-
tifoams makes it difficult to compare antifoam efficiency rigorously. In ad-
dition, the variety in stock formulations and equipment means that the results
of such a comparison in a mill can often be applied with certainty only to
166 Allen et al.

TABLE 5 Selected Kraft Browns tock Washer /Black Liquor Antifoam Patents
(Hydrophobic Particle Type)

Patent Type of particles Components/comments

F. J. Boylan (to Her- Hydrophobic silica Silica treated with silicone


cules Inc.), 1963 [15] Not exclusively for pulp and
paper
W. R. Christian and R. Hydrophobic silica Silica treated with amide, then
Liebling (to Nopco with silicone
Chemical Co.), 1965 Does not require additional
[18 J spreading agents
E. Domba (to Nalco Hydrophobic silica Silica treated with cyclic siloxane
Chemical Co.), 1968
[l 9 J
H. Lieberman, C. A. Aluminum oxide May contain silicone as spread-
Duharte-Francia, and J. ing agent
W. Henderson (to Betz May contain up to 10% water
Laboratories Inc.), 1972
[20]
T. F. MacDonnell (to Amide wax Bisstearamide wax
Diamond Shamrock
Corp.), 1972 L16]
J. H. Curtis and F. W. Amide wax Contains silicone
Woodward (to Diamond Talc Contains oil-soluble copolymer
Shamrock Corp.), 1972 of vinyl acetate and fumaric acid
[21]
I. A. Lichtman and A. Amide wax Amide of tallow fatty acid
M. Roscngart (to Dia- Contains silicone
mond Shamrock Corp.), Contains copolymer of vinyl ace-
1972, 1973 [22] tate and fumaric acid esterified
with tallow alcohol
Contains triglycerides to make
wax particles more amorphous
Covers water-extended and
water-based formulations
T. F. MacDonnell (to Hydrocarbon and/ May contain silicone
Diamond Shamrock or amide wax Contains copolymer of vinyl ace-
Corp.), 1972 (23] tate and fumaric acid
Covers water-extended and
water-based formulations
H. J. S. Shane, J. E. Amide wax Bisstearamide wax
Schill, and J. W. Lilley Contains silicone
(to Hart Chemical
Ltd.), 1973 [24]
Defoaming in the Pulp and Paper Industry 167

TABLE 5 Continued

Patent Type of particles Components/comments

R. J. Michalski and C. Amide wax Beheramide wax


C. Cochrane (to Nalco Contains copolymer of vinyl ace-
Chemical Co.), 1975 tate and ethylene
[25]
R. J. Michalski and R. Amide wax Methylene bisstearamide
W. Youngs (to Nalco Hydrophobic silica Contains silicone
Chemical Co.), 1975 Silica added to reduce viscosity
[26]
H. Lieberman, A. J. Amide wax
Graffeo, and J. S. Kuc-
san (to Betz Laborato-
ries Inc.), 1976 [27]
J. V. Sinka and I. A. Hydrophobic silica Amide wax of tallow fatty acid
Lichtman (to Diamond Amide wax added to stabilize antifoam
Shamrock Corp.), 1977
[28]
R. K. Berg and D. S. Amide wax Fatty acid diamide
Smalley (to Associated Hydrophobic silica Contains stearyl alcohol ethoxy-
Chemical Inc.), 1977 late emulsifier
[29] Water-based (oil-in-water
emulsion)
H. J. S. Shane and F. Amide wax Ethylene bisstearamide
S. Schell (to Hart Contains silicone
Chemical Ltd.), 1978 Water-based (oil-in-water
[30] emulsion)
DIC Hercules Inc., Amide wax Ethylene bisstearamide
1983 [31] Hydrophobic silica Contains silicone
Contains ethylene glycol for low
temperature stability
Water-extended
B. Danner (to Sandox Amide wax Ethylene bisstearamide
GmbH), 1983 [32] Hydrophobic silica Contains vegetable or animal oil
instead of mineral oil
Contains no silicone
Water-based (oil-in-water
emulsion)
168 Allen et al.

that particular mill, under the prevailing operating conditions, and are thus
difficult to evaluate.
In some mill studies, water-extended brownstock washer antifoams are
claimed to be superior to their oil-based analogs [33-35]. Viscosities are
similar to those of oil-based antifoams, so that the same pumping and dis-
pensing equipment can be used. The cost per unit weight is less, and benefits
reported include
Lower antifoam use
Better washing
Reduced biological oxygen demand in effluent
Reduced deposition of antifoam on process equipment.
Costs savings of several hundred thousand dollars per year per mill have
been claimed [33,34]. In the authors' experience, however, few mills realize
such benefits; at present, oil-based brownstock antifoams are still usually
considered the most cost effective.
At least one company [36,37] has a water-based brownstock washer an-
tifoam which it claims is as cost-effective as oil-based antifoams, without
many of their drawbacks. Water-based antifoams have, as yet, met with only
limited acceptance.
Water-based/hydrocarbon oil-free antifoams similar to the original sili-
cone emulsion antifoams that appeared in the 1950s are attempting a come-
back. They circumvent the problems involved with mineral oil; however,
the relatively expensive silicone renders them less cost-effective than the
standard oil-based antifoams.

C. Other Pulp and Paper Antifoams


Other types of antifoams are used in pulp and paper production where con-
ditions of temperature and pH are less extreme. The old paste-type antifoams
(high-viscosity oil-in-water emulsions of fatty acids, esters, waxes, etc.) lost
popularity largely because of handling problems, but, with the increase in
oil prices in the 1970s and improvements in pumping and metering equip-
ment, they have made something of a comeback. Blends of fatty acids, fatty
alcohols, and/or polyethylene glycols, in mineral oil, are easier to handle,
but have lost ground because of problems associated with the oil (see Sec.
V. D.). Some antifoams are sold without carrier liquids, as solids or un-
diluted liquids.
Most modem antifoams are pumpable fluids. Antifoams that have no car-
rier liquid save on transportation costs and are relatively stable when stored
at low temperatures ( <0°C); when added to the stock undiluted, they require
agitation to ensure thorough mixing. These carrier-free antifoams are usually
Defoaming in the Pulp and Paper Industry 169

fatty acids, fatty alcohols, and their derivatives, plus emulsifiers. Some con-
tain flammable components, and must be handled appropriately.
Aqueous emulsion antifoams (typically of long-chain fatty acids, fatty
alcohols, and esters) are gaining in popularity for several reasons. Water as
a carrier is less expensive than oil. High concentrations of active ingredients
(up to 50%, usually -30%, versus ~10% for oil-based antifoams) can be
tolerated before viscosity is a problem. Dispersion in stock is easier because
the defoaming material is already emulsified. Problems associated with oil
are reduced. In addition, on the paper machine they often appear to be more
efficient at defoaming than oil-based antifoams [38].
Some aqueous dispersions of relatively hydrophilic fatty acids and al-
cohols incorporate water in the droplets; in recent years, these have been
called "gel-particle" antifoams. Because the gel particles are hydrophilic,
they function somewhat differently from the hydrophobic particles in a
brownstock washer antifoam [39 ,40]. Apparently, the gel particle reacts with
polyvalent cations in the stock (e.g., calcium, magnesium, aluminum) to
form a hydrophobic shell around the particle, thus permitting it to enter the
film of a foam bubble. When watched in a microscope the particle appears
to "explode," disrupting the film and causing collapse or coalescence. Pat-
ents specifically for gel-particle antifoams (e.g., [41,42]) include aluminum
salts in the formulation; the hydrophobic shells thus are formed around the
gel particles before the anti foam is added to the process liquid.
Paper coating compositions resemble paints, and can use similar anti-
foams. Many pulp and paper antifoams can also be used in coatings, but
very hydrophobic formulations should be avoided, as they can cause spots
on the coated sheet.

D. Disadvantages of Chemical Defoaming

There are limitations and problems associated with the use of chemical
antifoams; the problems are frequently aggravated by high dosages of the
antifoams. With care, these difficulties can be minimized.
For high temperature applications, industry opinion favors oil-based or
sometimes oil-based/water-extended antifoams. The tendency of these an-
tifoams to deposit on process equipment [43], however, is a serious draw-
back to their use; the amount of deposition increases with dosage [44]. In
addition, overuse of oil-containing antifoams has been associated with re-
duced strength, brightness, uniformity, and opacity of the finished paper,
and with flaws in coating [7 ,45]. They should be used as far upstream in
170 Allen et al.

the pulp mill as possible, to maximize the washing of the antifoam from the
pulp.
Hydrocarbon oils in some brownstock washer antifoams have been linked
to trace quantities of polychlorinated dioxins and furans in products and
effluents from bleached kraft mills [46]. This may be avoided by the use of
oils free of the unchlorinatcd precursors of these toxins [47].
Kerosene and fuel oil share the deposition problem of oil-based anti-
foams. In addition, they are flammable, work only on surface foam, cause
deterioration of rubber parts, and give off odor in use and in the final product.
Oil-free antifoams avoid these problems. However, water-based emul-
sions are sensitive to temperature extremes and generally have a shorter life-
time than oil-based or carrier-free antifoams. Emulsifiable, carrier-free
antifoams are relatively stable, but require care when being dispersed to
achieve consistent results.

VI. MEASUREMENT OF FOAM


Monitoring of surface foam levels is relatively straightforward. With open
vats and screens, it can be done by eye; however, continuous monitoring
with electronic or pneumatic sensors is more precise.
Various systems have been developed to measure entrained foam in pulp
and paper stock; most are based on a measurement of the compressibility
of the stock. For example, in one such device f48,49] a stock sample is
isolated in an onstream sampling cylinder, then compressed with a test pis-
ton; the volume of compression and the pressure developed are fed into a
microprocessor whose output governs antifoam feed. Because large, free-
floating air bubbles contribute proportionately more to the measured en-
trained air volume, but often cause less trouble, than the smaller bubbles
adhering to solid particles in the stock, the amount of entrained air that can
be tolerated will depend on the size distribution of the bubbles. The maxi-
mum tolerance must be determined empirically for each mill, each site, and
each stock/antifoam combination. Thus, while average tolerable entrained
air limits are around 6%, actual limits vary from 2.5% to 10% [49].
A recent Finnish development [50] circumvents this by making use of
the fact that ultrasonic transmissions through water are attenuated consid-
erably more by small bubbles than by large bubbles. Initial trials look prom-
ising; the ultrasound device appears to react rapidly and precisely to trou-
blesome changes in the amount of entrained air.
To determine the total amount of foam (surface and entrained) in a stock
vessel, a system has been described [51,52] in which the measured level of
the air-liquid or air-foam surface is compared with the theoretical level of
unfoamed liquid calculated from measurements of the hydrostatic head at
Defoaming in the Pulp and Paper Industry 171

the bottom of the vessel. The difference gives a measure of the combined
volumes of surface and entrained foam. If the depth of the surface foam is
also measured, the amount of entrained air can be estimated.

VII. TESTING OF ANTIFOAMS


The suitability of an antifoam must ultimately be determined by mill trials,
but there are several laboratory tests that can be used for preliminary screen-
ing. Most measure only surface foam, which may not reflect the effect of
the antifoam on entrained air [49).
Typically, a quantity of stock, or stock plus antifoam, is put in a grad-
uated cylinder. Air is introduced into the stock in the cylinder by sparging
(bubbling controlled amounts of air into the stock through a frit in the bottom
of the cylinder) [53-55), by cascading one aliquot of stock into another
[56,57), by shaking the cylinder reproducibly [58], or by cascading the stock
with a recirculating pump [14). The amount of surface foam produced is
measured and correlated with the presence of antifoam.
The same tests can also be used to screen the efficiencies of antifoams
in paper coating formulations. In addition, a measure of the tendency of an
antifoam to cause defects in the finished coating may be obtained by spread-
ing a sample of the liquid coating plus antifoam on a clean glass plate,
drying the mixture, and examining the resultant film for imperfections [59).
Antifoams may be screened in the laboratory for their tendency to con-
tribute to pitch deposits by measuring deposition with a Vibromixer™ [60).
A new procedure now exists [44) which enables the antifoam to be tested
in pulp stock rather than in water, which is an advantage as the presence of
other components of the stock affects deposition rates [61). If deposits have
already formed in the mill, the amounts of antifoam components in the pitch
can be analyzed [43).

VIII. HANDLING AND USE OF ANTIFOAMS


The manufacturer of an antifoam should be consulted for specific instruc-
tions about its storage and handling. In general, antifoams should not be
exposed to temperature extremes. Water-containing antifoams are particu-
larly sensitive to temperature, and should be kept between 5°C and 20°C.
The storage life is a few months, but may be extended somewhat if the
antifoam is stirred intermittently during storage; care should be taken to avoid
entraining air during stirring.
Oil-based and carrier-free antifoams are more resistant to high and low
temperature; however, extremes should still be avoided. Some carrier-free
172 Allen et al.

antifoams become viscous and hard to handle at low temperatures ( <0°C).


Antifoams containing volatile or flammable materials should be stored below
30°C. Carrier-free antifoams usually have storage lives of about a year.
Emulsion and dispersion antifoams, and most other pumpable liquid
antifoams, may be added to the stock without dilution. Viscous or carrier-
free antifoams require vigorous agitation to ensure complete dispersion when
they are added to stock or diluted prior to use. Any diluted antifoam should
be stirred frequently until used. During any agitation of the stock or anti-
foam, air entrainment should be avoided.
Traditionally, antifoam was added by an operator when surface foaming
got out of hand. This is still done when necessary, but it is now customary
to add antifoam continuously to the stock by metering pumps; positive dis-
placement pumps with large diameter suction lines are recommended. An-
tifoams should be added to the stock at points where there is sufficient ag-
itation or turbulence for complete mixing (e.g., into flowing stock lines or
stirred vessels; before pumps). When the pulp forms a mat or web (e.g., on
washers), diluted antifoam may be sprayed on. Each antifoam addition site
should have its own pump and control.
Antifoam is metered into the stock at a constant rate with adjustments
made as needed when the amount of foam changes. Some form of automatic
foam sensing device is preferable to visual inspection (particularly if it gives
a measure of entrained air); the sensor location may have to be determined
by trial and error to optimize the correlation of sensor output with required
antifoam dose.
The antifoam feed rate can be adjusted manually, in response to change
in the amount of foam present. If frequent adjustments are necessary, how-
ever, it may be more economical to invest in an automated control system.
Claims have been made [48,49,51,62] of antifoam savings of from 20% to
50% when manual control of antifoam feed to a sensor-monitored stock sys-
tem was replaced with computerized control responding automatically to sensor
output. This is not surprising; a busy operator may make only periodic in-
spections of sensor output, while the computer monitors sensor output con-
tinuously, responding immediately to higher foam levels with a minimum
increase in antifoam addition rate before things get too bad, then decreasing
antifoam feed as soon as possible.
With experience, the need for changes in antifoam dosages can sometimes
be anticipated. Use of larger quantities of very resinous wood (e.g., pine,
aspen) or recycled process liquids, or less aging of the chips (shorter time
and/or freezing temperatures) will cause more foam, especially in the early
stages of the pulp mill. Colder makeup water in the winter and increased
amounts of dissolved or suspended material in the water during spring runoff
may also have a predictable effect on foaming. Foam levels should be watched
Defoaming in the Pulp and Paper Industry 173

carefully when physical or chemical changes are made in the system, or


when the type or brand of antifoam is changed.

IX. FUTURE RESEARCH AND DEVELOPMENT


The biggest developments in physical control of foaming in recent years
have been in the areas of kraft brownstock washing and screening. Dis-
placement washing, diffusion washers, horizontal pressure washers, pres-
sure knotters and screens, and more gradual release of digester pressure have
reduced greatly the amount of air entrained in the brownstock. The potential
for savings from improved washing and reduced foam is so great that con-
tinued progress in these areas can be expected for some time to come.
Similarly, in the field of chemical defoaming, it is the brownstock
antifoams that are attracting the most attention. Current oil-based antifoams
are efficient at defoaming, but the oil causes numerous problems; the oil-
free antifoams available presently are more expensive. Much research re-
mains to be done in the development of oil-free brownstock antifoams that
are both effective and inexpensive.
The industry has come a long way from the days of cream and kerosene.
No doubt the challenge of foam control in pulp and paper mills will continue
to be met creatively and effectively.

REFERENCES
1. Anonymous, Pulp Paper J. 42(1): 27 (1989).
2. Corpus Information Services, Market Study: Chemicals for Pulp and Paper in
Canada, Southam Communications, Ltd., 1985.
3. E. Sjostrom, J. Appl. Polym. Sci.: Appl. Polym. Symp. 37: 577 (1983).
4. G. A. Smook, Handbook/or Pulp and Paper Technologists, p. 15. Joint Text-
book Committee of the Pulp and Paper Industry (TAPPI/CPPA), 1982.
5. E. Sjostrom, lnt. Symp. Wood and Pulping Chem. (Japan) 3: 57 (1983).
6. L. H. Allen and M. C. Lapointe, Nordic Pulp Paper Res. J. 4: 94 (1989); 4:
229 (1989).
7. K. Minakuchi, Japan Pulp Paper 4(3): (1966).
8. H. Guo and B. A. W. Coller, Tappi J. 72(1): 50 (1989).
9. J. B. Ball and S. R. Forster, Proc. CPPA Western Branch Meeting, Jasper,
1984.
10. J. M. Macleod and M. E. Cyr (assigned to Paprican and Irving Pulp and
Paper Ltd.), US 4,814,041; March 21, 1989; filed March 18, 1987.
11. Anonymous, Pulp Paper J. 38(3): 22 (1985).
12. K. A. Coffey, Paper Trade J. 169(1): 67 (1985).
13. J. P. Campbell, Pulp Paper 56(7): 121 (1982).
14. I. Lichtman and T. Gammon, in Kirk-Othmcr Encyclopedia of Chemical Tech-
nology, Vol. 7, 3rd ed., Wiley, New York, 1979, pp 430-448.
174 Allen et al.

15. F. 1. Boylan (assigned to Hercules Inc.), US 3,076,768; February 5, 1963;


filed April 5, 1960; Also: Can 662,736; May 7, 1963; filed July 6, 1960.
16. T. F. MacDonnell (assigned to Diamond Shamrock Corp.), US 3,652,453;
March 28, 1972; filed January 27, 1970.
17. R.H. Pelton, Pulp Paper Canada 90(2): T6l (1989).
18. W. R. Christian and R. Liebling (assigned to Nopco Chemical Co.), US
3 ,207 ,655; September 21 , l 965; filed February 13, 1963.
19. E. Domba (assigned to Nalco Chemical Co.), US 3,388,073; June 11, 1968;
filed December 12, 1963.
20. H. Lieberman, C. A. Duharte-Francia, and J. W. Henderson (assigned to Betz
Laboratories Inc.), GB 1,267,479; March 22, 1972; filed May 22, 1970.
21. J. H. Curtis and F. W. Woodward (assigned to Diamond Shamrock Corp.),
US 3,673,105; June 27, 1972; filed April 11, 1969.
22. I. A. Lichtman and A. M. Rosengart (assigned to Diamond Shamrock Corp.),
US 3,677,963; July 18, 1972; filed September 30, 1970. Also: Can. 927,707;
June 5, 1973; filed September 3, 1971.
23. T. F. MacDonnell (assigned to Diamond Shamrock Corp.), Can. 909,630;
September 12, 1972; filed August 18, 1971.
24. H. J. S. Shane, J. E. Schill, and J. W. Lilley (assigned to Hart Chemical
Ltd.), Can. 922,456; March 13, 1973; filed December 31, 1971.
25. R. J. Michalski and C. C. Cochrane (assigned to Nalco Chemical Co.), US
3,893,941; July 8, 1975; filed September 9, 1971.
26. R. J. Michalski and R. W. Youngs (assigned to Nalco Chemical Co.), Can.
971,851; July 29, 1975; filed June 5, 1972. Also: US 3,923,683; December
2, 1975; filed September 7, 1971.
27. H. Lieberman, A. J. Graffeo, and J. S. Kucsan (assigned to Betz Laboratories
Inc.), US 3,935,121; January 27, 1976; filed June 29, 1973.
28. J. V. Sinka and I. A. Lichtman (assigned to Diamond Shamrock Corp.), US
4,021,365; May 3, 1977; filed January 24, 1973.
29. R. K. Berg and D. S. Smalley (assigned to Associated Chemists Inc.), US
4,032,473; June 28, 1977; filed June 4, 1975.
30. H.J. S. Shane and F. S. Schell (assigned to Hart Chemical Ltd.), US 4,088,601;
May 9, 1978; filed November 11, 1974.
31. DIC Hercules Inc., Jap. 57 24,309 [83 24,309]; February 14, 1983; filed Au-
gust 7, 1981.
32. B. Danner (assigned to Sandoz GmbH), DE 3,242,202; June 1, 1983; filed
November 25, 1981.
33. R. T. Maher, Proc. Tappi Pulping Conf., p. 461, 1981.
34. F. Poltenson, Pulp Paper 60(2): 97 (1986).
35. L. F. Twoomey, Proc. Tappi Papermakers Conf. (Atlanta), p. 241, 1984.
36. L. F. Twoomey, Southern Pulp Paper 49(5): 33 (1986).
37. Nalco Chemical Co. Product Information.
38. R. H. Lorz, Pulp Paper Canada 88(10): 85 (1987).
39. R. C. Montani and F. J. Boylan, Proc. Tappi Annual Meeting, p. 293, 1980.
40. R. C. Montani and F. J. Boylan, Proc. Tappi Pulping Conf., p. 49, 1982.
Defoaming in the Pulp and Paper Industry 175

41. F. J. Boylan (assigned to Drew Chemical Corp.), US 4,303,549; December


l, 1981; filed October 18, 1979.
42. F. J. Boylan (assigned to Drew Chemical Corp.), US 4,340,500; July 20,
1982; filed October 18, 1979.
43. G. M. Dorris, M. Douek, and L. H. Allen, J. Pulp Paper Sci. 11(5): 149
(1985).
44. N. Dunlop-Jones and L. H. Allen, J. Pulp Paper Sci. 15(6): 235 (1989).
45. A. M. Springer, J. P. Dullforce, and T. H. Wegner, Tappi J. 69(4): 106
(1986).
46. L. H. Allen, R. M. Berry, B. I. Fleming, C. E. Luthe, and R. H. Voss,
Chemosphere 19: 741 (1989).
47. R.H. Voss, C. E. Luthe, B. I. Fleming, R. M. Berry, and L. H. Allen, Pulp
Paper Canada 89(12): 151 (1988).
48. J. Morton, Trans. PIMA Meeting (Seattle), 1986.
49. S. J. Dougherty, Tappi J. 72(1): 50 (1989).
50. M. Karras, T. Pietikiiinen, H. Kortelainen, and J. Tomberg, Tappi J. 71(1):
65 ( 1988).
51. G. L. Cooper, Proc. Tappi Annual Meeting (Atlanta), p. 71, 1988.
52. W. E. Sande, Tappi J. 69(3): 246 (1986).
53. ASTM Method 01881-86 (1986).
54. W. L. Carpenter and I. Gellman, Tappi J. 50(5): 83A (1967).
55. M. Penttila, T. Laxen, and N.-E. Virkola, Pulp Paper Canada 87(1): 82 (1986).
56. ASTM Method DJ 173-80 (1986).
57. I. YrjiiUi, R. Rasanen, and H. Bruun, Paperija Puu 46(4a): 257 (1964).
58. H. Hollaender, Paper-Maker (London) 130: 564 (1955).
59. Sandoz Chemicals Corp. Product Information.
60. Tappi Useful Method 223.
61. M. Douek and L. H. Allen, J. Pulp Paper Sci., Trans. Tech. Sect. CPPA
9(2): TR48 (1983).
62. M. Andrews, Pulp Paper 56(2): 153 (1982).
4
Application of Antifoams in
Pharmaceuticals

ROLLAND BERGER Th. Goldschmidt AG, Chemische Fabriken,


Essen, Germany

I. Introduction 177
II. Silicone Antifoam Materials and Their Toxicity 179
III. Regulations 181
IV. Formulations 184
A. Tablets 184
B. Suspensions 185
C. Emulsions 185
V. In Vitro and In Vivo Experiments 186
VI. Conclusions 189
References 189

I. INTRODUCTION
Pharmaceutical applications of polysiloxane (silicone) chemistry staited in
the early fifties, when polydimethylsiloxanes were first used as antifoams
in the manufacture of drugs [ 1]. Since this introduction more than 35 years
ago, an increasing number of fields of application have emerged [2- 7]. One
of the earliest applications of polysiloxanes, which is still of great interest,
is their use in pharmaceuticals to combat intestinal gas.
The first therapeutic application of polysiloxanes against digestive upsets
was done by Quin et al. [8], who examined the flatulence in ruminants.
Hirschowitz et al. [9], Gasster et al. [10], and Rider et al. [11-13] later
reported the successful application of polysiloxane-based antifoams in the
177
178 Berger

prevention of intestinal foam during gastroscopy, which otherwise hindered


the inspection of the gastrointensial tract [ 14].
Intestinal gas, entrapped in foam bubbles, may lead to any form of in-
digestion, discomfort, fullness, and pressure or pain in the gastrointenstinal
tract accompanied by lack of appetite and vitality or general fatigue. These
complaints are normally referred to as dyspepsia. The main indications of
dyspepsia are hyperacidity and flatulence, which frequently occur together.
The intestinal gas is mainly composed (99%) of nitrogen, oxygen, hydrogen,
carbon dioxide, and methane, and the distribution of these gases may vary
over a wide range [15,16].
Dyspepsia may result from different causes [ 17]. Air swallowing is rec-
ognized to be one of the most common sources of abdominal distension,
especially by those people who eat and drink too fast, smoke, or are emo-
tionally disturbed. Swallowed air can contribute up to 70% of the gastroin-
testinal gas [ 18].
Other causes of gastrointestinal gas are certain foods which enable the
colonic bacteria to be particularly productive in the formation of gas. The
substrate in the intensine and the predominant organism will influence the
flatus, and the volume will depend on the quantity of substrate acted upon
by the bacteria. Steggerda [ 19] has reported that the volume of flatus can
be increased more than tenfold by ingestion of beans. The gas produced is
mainly carbon dioxode.
Additionally a deficiency in digestive function may result in gastrointes-
tinal malfunction. This can be caused either by reduced digestive enzymes
or stress, an activity which is known to increase the production of mucus
and to increase gastrointestinal motility, entrapping whatever gas is present
in the bowel. A deficiency of pancreatic enzymes has been demonstrated to
be one of the physiological changes of aging. So gastrointestinal complaints
are more common in the elderly [20,21].
Stress is also known to lead to inflammation of the gastrointestinal tract,
thus disturbing the absorption and the diffusion of intestinal gas into the
blood. Gas therefore cannot be eliminated via the lungs.
As pointed out, flatulence is frequently accompanied by hyperacidity. A
study by Graham et al. [22] reported that 50% of the American population
have used an antacid at some time in their lives and that 27% take two or
more doses a week. Seventy-five percent of the regular users take six or
more doses every week. Dyspepsia therefore seems to be a very common
complaint.
The described causes of dyspepsia have given rise to pharmaceuticals to
combat the digestive upset. The pharmaceuticals normally contain anti-
foams, antacids, and/or enzymes.
Antifoams in Pharmaceuticals 179

II. SILICONE ANTIFOAM MATERIALS AND THEIR


TOXICITY
Polysiloxanes are polymers with repeating units of organosiloxy groups with
general formula R,,Si0( 4 _ nJ/ 2 with a varying degree of substitution, and
therefore varying functionality. The four basic structures of the repeating
units and their generally used abbreviations are
R R
I I
R-Si--0 1; 2 01;2-Si--01;2
I I
R R
M (n = 3) D (n = 2)
R 01;2
I I
0 112 -Si --01;2 01;2-Si--01;2
I I
01;2 01;2
T (n = l) Q (n = 0)

These repeating units in the polymers are derived from the corresponding
halosilanes R,,SiX( 4 - nJ by hydrolysis under the elimination of HCI. A mixture
of the halosilanes is produced from elementary silicon and organic halides
RX using the Rochow synthesis, followed by a distillation process to yield
the pure monomers. Examples of the organic halides are especially CH3Cl
and in part C 6H 5 Cl, leading to methylchlorosilanes or phenyl group con-
taining chlorosilanes.
Depending on the type and on the ratio of the chlorosilanes used during
the hydrolysis, cyclic, linear, branched or cross-linked oligomers and poly-
mers are achieved [23]. Using water in excess, the polymers contain OH
end groups, with the exception of the cyclic compounds. In a further re-
action, the so-called equilibration reaction, the polymers can be tailored with
respect to their molecular weight, for which the viscosity of the products is
a measurement. The relationship between the molecular weight of the (lin-
ear) polymers and the viscosity can be described by the empirical equation
log lJ (cSt/25°C) = 1.00 + 0.0123M 05
where lJ is the viscosity and M is the molecular weight. This equation was
developed by Barry [24] and is valid for M > 2500. Other equations have
been developed by Piccoli and Stark [25] and Weissler [26]. The equilibra-
tion reaction is an acid- or basically catalyzed random rearrangement of sil-
180 Berger

oxy bonds providing a statistical distribution of molecular sizes. Products in


the viscosity range of about 20 mPa s up to 106 mPa s are commercially
available. As chain-limiting end groups, M units can be added. The equi-
libration process is characterized by the formation of a constant amount of
cyclic oligomers with octamethylcyclotetrasiloxane (D 4 ) as the main com-
pound of the cyclic. The low molecular weight polysiloxanes are then re-
moved by distillation and the residual polymer mixture can be further pu-
rified when used for medical applications. It has been shown that higher
contents of volatile, low molecular weight polysiloxanes lead to an increased
appearance in the urine of humans, while other experiments have shown that
polysiloxanes with molecular weights higher than 500 were poorly absorbed
from the gastrointestinal tract [27].
For use in pharmaceuticals the linear polymers of the general formula
(MDxM)
with R = CH3 (polydimethylsiloxane) are well established and referred to
as dimethicone in the pharmaceutical literature. Dimethicone is an ideal raw
material for antifoam formulations due to its low surface tension. The sur-
face tension has values of about 20-22 mN m- 1 , depending on the molecular
weight (for the dimeric compound hexamethyldisiloxane (M-M) the sur-
face tension decreases to about 16 mN m- 1). The low surface tension is an
important property of antifoam raw materials which is essential for their
mode of action [28-34]; see Chapter 1.
When used alone, dimethicones are ineffective antifoams. They gain their
extraordinary antifoam properties when formulated with finely divided hy-
drophobic solids. These solids can in principle be organic [35-37] or in-
organic. For application in pharmaceuticals to combat flatulence highly dis-
persed hydrophobed silica is commonly used. Silica is comprised of spherical
particles with a mean particle size of about 12-15 nm when it is of the fumed
variety. Other types of silica are precipitated from water glass but are not
preferred for the formulation of pharmaceutical antifoams because they can
contain excess alkali. Commercially available silica is surface-treated to yield
hydrophobic types. The treatment can be done by the reaction of the OH
groups at the surface with, for example, (CH 3hSiCl or with polydimethyl-
siloxanes under the influence of heat and/or catalysts and mechanical shear
forces. Treatment of the silica with organic substances is also known. Mix-
tures of dimethicone with silica are normally referred to as simethicone.
Simethicone is either produced by mixing pretreated silica with dimethicone
or by mixing untreated silica with dimethicone followed by an in situ treat-
ment at elevated temperatures [38]. A mixture of untreated silica with di-
methicone is characterized by a high viscosity which is shear sensitive. This
means that the viscosity of such a mixture is strongly reduced under the
Antifoams in Pharmaceuticals 181

influence of shear forces (thixotropoy). This effect is caused by interparticle


action of the silica which is reduced under shear. The same mixture with
an equal quantity of pretreated or in situ treated silica is less viscous and
shows a smaller thixotropic effect. It has, however, been shown that si-
methicones containing pretreated or in situ treated silica are more effective
as antifoams than those with untreated silica [39 ,40]. The role of the silica
in simethicone antifoams is discussed elsewhere [31,41-43]. A detailed ac-
count is found in Chapter 1 .
Besides the physicochemical properties which make dimethicone an ideal
raw material for antifoams, their nontoxicity make dimethicones ideal for
use in pharmaceuticals, especially for those which are designed for oral use.
Dimethicones are recognized as chemically inert and physiologically neu-
tral. Early toxicological studies including dimethicone were conducted by
Rowe et al. [44]. These studies indicated that commercially available di-
methicone maintained in stock diets for a group of female rats at levels up
to 1% for three months showed no significant abnormalities in the liver,
spleen, kidneys, heart, lungs, pancreas, or stomach. The studies were ex-
tended to male and female rats by feeding 0.3% simethicone for two years
showing no adverse effects on growth, mortality, or behavior [45]. During
subacute studies with simethicone fed to dogs in doses from 0.3 g/kg up to
3 g/kg no symptoms of toxicity were noted [46).
Only mild effects of moist, loose stools were observed. The list of tox-
icological studies can be completed by the studies of Gloxhuber and Hecht
[47], who carried out acute and chronic oral as well as intraperitonal, in-
travenous, and dermal sensitivity investigations on rats, cats, and rabbits,
using dimethicone of different viscosities. In no case was any toxic action
observed. McDonald et al. [48] conducted subacute oral toxicity tests using
dimethicone of 50 up to 60,000 cSt at a level of 1% in the feedstock for
rats over 90 days. Leucocyte counts and haemoglobin were normal. The
effects of dimethicone on reproduction and fetal development in rats and
rabbits and the mutagenic potential in mice were evaluated by Kennedy et
al. [49]. No evidence of fetotoxicity and mutagenic effects were obtained.
Cutler et al. [50] ran long-term studies with simethicone incorporated in the
diet of mice at levels of 0.25% to 2.5% for 76 weeks. No significant his-
topathologic changes were observed. Single doses of silicones were given
to rabbits, cats, dogs, and rats as high as 54 mL/kg. Except for diarrhea
there were no indications of negative effects [51,52]. Silicone fluids were
excreted unchanged and there was no absorption [53].

Ill. REGULATIONS
Because of their inertness and physiologically neutral behavior, simethi-
cones are admitted as antiflatulents for "over the counter human use" by the
182 Berger

TABLE 1 Essential Specification for Dimethicone According to National Formu-


lary [55]

Viscosity (at 25°C): 18-13750 cSt (USP method No.911) [57]


Specificity gravity: 0.946-0.975 g/ml (USP method No.841) [58]
Refractive index: 1.3980-1.4055 (USP method No.831) [59]
Loss on heating (15 g; 200°C; 4 h): <2% (except Dimethicone 20 cSt: <20%)
Heavy metals: <0.001% (10 ppm) (USP method No.231,II) [60]
Source: Ref. 55.

United States Food and Drug Administration (21 CFR Part 332). The max-
imum daily dose is limited to 500 mg. The antiflatulent may also contain
"any generally recognized as safe and effective antiacid ingredient(s)" as
sanctioned under 21 CFR Part 331. Moreover, polydimethylsiloxanes and
silica are allowed as antifoam formulations in food and food processing (with
concentration limitations!) [54].
The specification for dimethicone, silica, and simethicone can be found
in the United States National Formulary [55] and the U.S. Pharmacopeia
[56], respectively. The National Formulary is a sister compendium to the
U.S. Pharmacopeia, describing pharmaceutical ingredients which can be used
in the formulation of pharmaceuticals, while the Pharmacopeia itself is lim-
ited to drug substances and dosage forms.
Following the National Formulary specification, dimethicone is
a mixture of fully methylated linear siloxane polymers containing re-
peating units of the formula

stabilized with trimethylsiloxy end-blocking units

wherein n has an average value such that the corresponding nominal vis-
cosity is in a discrete range between 20 and 12,500 centistokes (cSt). It
contains not less than 97. 0 percent and not more than 103. 0 percent of
polydimethylsiloxanc ([-(CH 3 hSi0-]n).
The essential specifications are described in Table I. Silica is referred to
as colloidal silicon dioxide in the National Formulary and defined as
submicroscopic fumed silica prepared by the vapor phase hydrolysis of a
silicon compound. When ignited at 1000° for 2 hours it contains not less
than 99.0 percent and not more than 100.5 percent of Si02 .
The specifications are summerised in Table 2.
Antifoams in Pharmaceuticals 183

TABLE 2 Specification for Colloidal Silicon Dioxide According to National For-


mulary [55)
pH value: 3.5-4.4
Loss on drying: <2.5%
Loss on ignition: <2%
Content of arsenic: <8 ppm [611
Source: Ref. 55.

According to the U.S. Pharmacopeia [56], simethicone is


a mixture of fully methylated linear siloxane polymers containing re-
peating units of the formula
[-(CH 3)zSi0-1n
stabilized with trimethylsiloxy end-blocking units of the formula

and silicon dioxide. It contains not less than 90.5 percent and not more
than 99.0 percent of polydimethylsiloxane ([-(CH 3 )zSi0-Jn), and not
less than 4.0 percent and not more than 7.0 percent of silicon dioxide.
It should be noted that the SiO" content is restricted, while simethicone
formulations for other than pharmaceutical applications may contain Jess or
more Si02 .
Simethicone is mainly specified [56] by the data given in Table 3. Similar
descriptions for simethicone do not exist in other countries, so the U.S.
Pharmacopeia definition [56] is normally accepted in many other states.
Specifications for dimethicone are also given in the United Kingdom
(Pharmaceutical Codex), France, and Germany (DAB) [66], but the de-
scriptions differ from those given in the U.S. National Formulary [55]. So,
for example, dimethicone (dimeticon) is specified in the DAB 9 [66] as

TABLE 3 Essential Specification of Simethicone according to US Pharmacopeia


[56]
Loss on heating (15 g; 200°C; 4 h): < 18%
[compared with the definition of NF XVII for dimethicone, simethicone can con-
tain a significantly higher amount of volatile, low molecular weight polymers]
Heavy metals: <0.001% (10 ppm) (USP method No.231,Il) (62]
Defoaming activity against octoxynol 9 [63] (l %)-solution using 20 ppm simethi-
cone: foam collapse time < 15 s
Source: Ref. 56.
184 Berger

TABLE 4 Specification for Dimethicone According to DAB 9 [66]

Consumption of NaOH (0.01 w) to estimate the content of acidic substances: <


0.15 mL
Viscosity (25°C): 20 - 1000 cST
Content of mineral oils: Compared with 0.1 ppm quinine sulphate in 0.01 N sul-
phuric acid
Heavy metals: <5 ppm
Loss of heating (1 g; 150°C; 2 h: <0.3% (for dimethicone < 50 cSt)
Source: Ref. 66.

summarized in Table 4. Even if there is no listing for simethicone in coun-


tries like Germany there are numerous pharmaceutical preparations avail-
able. All these formulations undergo a separate registration by the health
authorities. The so-called Rote Liste [67] summarizes pharmaceutical
antiflatulent preparations and describes the composition of the active ingre-
dients. Not all of these pharmaceuticals use basic dimethicone which cor-
responds to the DAB 9 description. Besides dimethicone with 500-cSt or
1000-cSt dimethicones with a viscosity of 2000 cSt or 3000 cSt are also
included.

IV. FORMULATIONS
Pharmaceuticals combating flatulence are offered in different dosage forms
[68-72]. Dosage forms like simethicone tablets, simethicone oral suspen-
sions, and simethicone emulsions are defined by the U.S. Pharmacopeia
[56). Simethicone tablets seem to be the most common dosage form.

A. Tablets
Tablets are formulated as pressed mixtures of simethicone with powdered
components like milk powder [73) or saccharides [74] where simethicone is
the only active ingredient. Simethicone tablets available on the market may
contain between 20 mg and 300 mg simethicone, using mainly 40 mg per
tablet.
According to the U.S. Pharmacopeia [56] simethicone tablets "contain
an amount of polydimethylsiloxane ([-(CH 3}zSi0-ln) that is not less than
85.0 percent and not more than 115.0 percent of the labeled amount of Si-
methicone." The defoaming activity, defined as foam collapse time of a
foamed octoxynol 9 ( 1% ) solution should not exceed 45 s using 200 ppm
simethicone.
In most pharmaceuticals, however, simethicones are combined with other
ingredients designed to overcome symptoms associated with dyspepsia. The
Antifoams in Pharmaceuticals 185

main additional ingredients are antacids such as different aluminum salts


(carbonates, hydroxides, phosphates, etc.), bismuth salts (aluminates, car-
bonates, subcarbonates, etc.), calcium salts (carbonates, phosphates, etc.),
magnesium salts (oxides, hydroxides, carbonates, trisilicates, aluminumsil-
icates, etc.), sodium and potassium salts (bicarbonates, carbonates, tartrates,
citrates, etc.) or combinations of these salts, because flatulence frequently
is accompanied by hyperacidity [75-85].
In many cases simethicones are also combined with different enzymes
especially lipases and/or proteases [62] to overcome gastrointensial mal-
function caused by reduced digestive enzymes.
Although simethicones are found to have an increased antifoam activity
when compared with dimethicone, there are numerous formulations avail-
able containing only dimethicone.

B. Suspensions
Simethicone oral suspension [56] is "a suspension of Simethicone in water.
It contains an amount of polydimethylsiloxane ([-(CH 3) 2Si0-]") that is not
less than 85.0 percent and not more than 115.0 percent of the labeled amount
of Simethicone."
The defoaming activity, defined as foam collapse time of a foamed oc-
toxynol 9 (1 % ) solution, should not exceed 45 s using 200 ppm simethicone.
Again suspensions can be formulated containing additional ingredients like
antiacids and/or enzymes.

C. Emulsions
Simethicone emulsion according to the U.S. Pharmacopeia [56] is
a water dispersible form of Simethicone composed of Simethicone, suit-
able emulsifiers, preservatives and water. It may contain suitable viscos-
ity-increasing agents. It contains an amount of polydimethylsiloxane
(-[(CH 3) 2Si0-1n) that is not less than 85.0 percent and not more than
110.0 percent of the labeled amount of Simethicone.
The defoaming activity, defined as foam collapse time of a foamed oc-
toxynol 9 (1 %) solution, should not exceed 15 s using 50 ppm of simethicone.
Suitable emulsifiers may for example be polysorbates with different amounts
of ethylene oxide, monoglycerides, and sorbitan esters with different fatty
acids, acetylated monoglycerides, or ethoxylated fatty alcohols and fatty acids
as defined in U.S. National Formulary [55]. These emulsifying agents also
comply with the U.S. Food and Drug Administration regulation [86] and
are recognized as safe in food ..
186 Berger

As viscosity-increasing agents agar, guar gum, starch xanthan gum, dif-


ferent cellulose derivatives as well as certain polyacrylates are listed in U.S.
National Formulary [55].
Preservatives are used to protect the organic-based emulsifiers and thick-
ening agents against biodegradation by microorganisms. Dimethicone and
simethicone themselves are not biodegradable. According to the U.S. Phar-
macopeia [56] the aerobic microbial counts should not exceed 100 per gram
[87]. In addition, as for the simethicone itself, the heavy-metal content of
simethicone emulsion should not be greater than 0.001 % (l 0 ppm).

V. IN VITRO AND IN VIVO EXPERIMENTS


The different amounts of simethicone used as a quality control and the dif-
ferences of maximum foam collapse time in the description of the U.S.
Pharmacopeia [56] indicate, that the efficiency of the pharmaceutical an-
tifoams will vary with their formulation. Unfortunately there is little infor-
mation about the methods of production of these pharmaceuticals. The
"quality" of the antifoams can therefore only be estimated by experimental
trials. Apart from the detailed description in the U.S. Pharmacopeia [56]
some other methods are described.
One of the most widely used in vitro test methods is that described by
Rezak [88]. This method is a shaking test using a sodium lauryl sulfate
solution (100 mL; cone. of surfactant: 2.5 gL _,) in a graduated cylinder
(volume, 250 mL). Using this method Cox and Nijland[89] have compared
antif1atulant preparations available in the Netherlands. The tests were carried
out by creating a certain foam volume (250 mL). Then the pharmaceuticals
were added in an amount corresponding to 40 mg simethicone. After mix-
ing, the foam volume and the foam collapse time were estimated. Cox and
Nijland found that the formulations containing simethicone as the only drug
are the most efficient. Preparations containing antiacids show reduced an-
tifoam properties.
Carless et al. [90] have proposed another test method during their inves-
tigation of the effect of finely divided silica on the antifoam properties. They
have introduced a dynamic foam test by aerating a stirred polyoxyethylene-
stearyl-cetyl ether solution through a sintered glass disk. After creation of
a certain foam volume the antifoam was added and the time measured until
complete foam disappearance occurred. Using these tests, Carless et al. [90]
showed that solvent extraction of dimethicone from simethiconc tablets led
to a considerably reduced efficiency of the tablets. Unfortunately Carless et
al. [90] gave no test results with the extracted dimethicone, so that their
conclusion that dimethicone is primarily responsible for the antifoam prop-
erties is in part contrary to test results, which have shown, that the pure
Antifoams in Pharmaceuticals 187

dimethicone is only a poor antifoam and that simethicone shows a signifi-


cantly enhanced efficiency [911. Nevertheless the results show, that di-
methicone, strongly absorbed on solid particles in the tablets as a thin film,
is a weak antifoam. This may lead to the conclusion, that simethicone for-
mulated with antacids to form a tablet may, due to the properties of di-
methicone, spread over the particles thus loosing antifoam efficiency. These
effects may be an explanation for the results found by Cox and Nijland [89].
Similar phenomenon are well known where simethicone antifoams are
used in powdered detergents. If the antifoams are not prevented from spread-
ing over the detergent powder (for example, by encapsulation) the efficiency
is lost after very short storage times. For this application antifoam formu-
lations have been developed where the polysiloxane is prevented from
spreading f92-95].
While the antifoam effect may be reduced when simethicone is formu-
lated with antacids the acid neutralization capacity of the antacids, on the
other hand, is not significantly changed [84,96].
Aeration tests have also been carried out by Stead et al. [97]. Here the
foaming medium constituted a solution containing hydroxyethylcellulose and
sodium lauryl sulfate which was adjusted to a pH value of 1.2 with hydro-
chloric acid. Stead et al. [97 J have compared three different antacid prep-
arations (partly containing simethicone) and also assigned a ranking in re-
spect to the antifoam properties. They have found that the combination of
hydrotalcite with simethicone showed a greater antifoaming effect than alu-
minum hydroxide/dimethicone did. This ranking was also confirmed by in
vivo tests with 12 patients.
A third method to test the efficiency of pharmaceutical simethicone prep-
arations is described by Loebel [98]. Using sodium lauryl sulfate as the
surfactant, a 1% solution is circulated with a flow inducer. The circulated
solution falls back into its reservoir, thus creating foam. Adjusting the flo-
wrate a constant foam volume is produced. Powdered simethicone tablets
are then added, and the foam volume is estimated after certain times of
circulation. The tests are repeated with standard gastric and gut juice cor-
responding to DAB 7 [66]. The results found, indicate, that for the different
foaming systems the ranking of the antifoams with respect to their efficiency
did change. Nevertheless, the most efficient preparation against sodium lau-
ryl sulfate was the best in the other media too. The differences found should
have been caused by the different formulations and the production technol-
ogy respectively.
The tests described above can be done in laboratories. Initial tests have
always been conducted using synthetic foam stabilizers, where sodium lauryl
sulfate is preferred. Unfortunately, extensive test series which are possible
in vitro using these tests cannot be done in vivo. So for in vivo tests usually
188 Berger

only one antiflatulence preparation is tested in a "double blind" study. Al-


though the efficiency of simethicone was established about 35 years ago,
there are numerous recent publications demonstrating the efficiency of new
formulations. Efficiency of the pharmaceuticals is normally determined by
the reduction of discomfort and pain and/or by sonografic, radiological, or
endoscopic examinations.
Gladisch et al. (99], for example, described the efficiency of two si-
methicone preparations of different formulations, used for premedication during
an abdominal ultrasound investigation with 40 patients. They have found
that both pharmaceuticals improved the quality of imaging of the abdominal
vessels and the epigastric structures. Similar results were found by Brock-
mann (100] during his sonographic examination and by Fixa et al. (101]
during their radiological examinations.
Konstantinidis [ 102] demonstrated improvements in endoscopy by using
a dimethicone preparation which permitted good assessment of the mucosa
of the stomach and the duodenum in about 90% of all cases. The investi-
gation has been done with 578 patients.
Significant improvement of the symptoms of dyspepsia has been found
in the treatment of 43 elderly female patients with an antacid/simethicone/
enzyme combination by Gotzes [103](reduction of symptoms, 87%), by Vi-
vat (104] using an aluminum hydroxide/magnesium hydroxide/cholestra-
mine/simethicone formulation (reduction of ulcer size; 80%; 28 patients),
and by Frilhsorger and Fuchs [ 105] applying aluminum sodium carbonate
dihydroxide/simethicone (improvement, 79%; 60 patients (male and female)).
Comparable results were obtained by Eveld (106] using an enzyme/si-
methicone combination (improvement, 80%; 39 patients; placebo, 38%; 40
patients), by Auld [107] treating 22 male and female patients with an en-
zyme/simethicone agent (patients' response to therapy, 100%) or by Schroe-
der [108] applying an enzyme/simethicone preparation (improvement, 100%;
placebo improvement, 21 %) .
Besides the reduction of symptoms caused by dyspepsia some authors
reported additional effects of simethicone during their examinations. Birtley
et al. [75] have found a protection of the mucosa against aspirin-induced
gastric irritation in rats, while Neumeyer and Rogozinski [109] observed
improvement from pain caused by duodenic ulcer ( 12 patients).
Simethicone is not only useful for combatting foam formation in the gas-
trointestinal tract, but it also improves the properties of double-contrast bar-
ium enema [ 110].
There have been a few attempts to use materials other than simethicone
as antifoams in pharmaceutical preparations. Davies et al. [111] have tested
lecithin as an antifoam agent. Compared with simethicone a 20-fold greater
amount has to be used to obtain comparable results.
Antifoams in Pharmaceuticals 189

VI. CONCLUSIONS
Silicone antifoams (simethicones) have been used in pharmaceuticals to combat
digestive upsets or to enhance the inspection of the gastrointestinal tract for
more than 35 years. They still play an important role today. This is dem-
onstrated by numerous publications over the years which have again and
again demonstrated the effectiveness of simethicone as an antifoam against
abdominal foams. Unfortunately, there is no information about the surface
active species in gastric juices which stabilize foam in the gastrointestinal
tract. Simethicones have gained their position due to their chemical inertness
and their neutral physiological behavior. It has to be noted that there are
regulations, which define the materials to be used in pharmaceutical appli-
cations. The U.S. Pharmacopeia in connection with its sister compendium
the National Formulary seems to be the most important, which is accepted
nearly everywhere.

REFERENCES
1. L. E. Casida, Industrial Microbiology, Wiley, New York, 1968.
2. A. Rembaum and M. Shen, Biomedical Polymers, Marcel Dekker, New York,
1971.
3. N. Kossovsky and J. P. Heggers, CRC Crit. Rev. Biocompat. 3:53 (1987).
4. H. W. Weissenburger and M. G. von der Hoeven, US 3499909, (1970).
5. A. F. Asker and C. W. Whitwirth, J. Pharm. Sci 63: 1630 (1974).
6. F. S. Rankin, Chimicaoggi, 37 (1986).
7. C. E. Creamer, Pharma. Technol., 79 (1982).
8. A. H. Quin, J. A. Austin, and K. Ratcliff, J. Am. Vet. Med. Assoc. 114:
313 (1949).
9. M. D. Hirschowitz, R. J. Bolt, and H. M. Pollardt, Gastroenterology 27:
649 (1954).
10. M. Gasster, J. 0. Westwater, and W. E. Molle, Gastroenterology 27: 652
(1954).
11. M. E. Dailey and J. A. Rider, J. Am. Med. Assoc. 155: 859 (1954).
12. J. A. Rider and H. C. Moeller, J. Am. Med. Assoc. 174: 2052 (1960).
13. J. A. Rider, Ann. N. Y. Acad. Sci. 150: 170 (1968).
14. E. Brunk, Diagnostika 9: 637 (1976).
15. I. A. D. Bouchier, Practitioner 224: 373 (1980).
16. M. M. van Nees and E. L. Cattau, Am. Fam. Physician 31: 198 (1985).
17. G. Trieb and E. Nusser Med. Welt 24: 1045 (1973).
18. J. Weiss, Curr. Therapeutic Res. 16: 909 (1980).
19. F. R. Steggerda and J. F. Dimmick, Am. J. Clin. Nutrit. 19: 120 (1966).
20. J. Weiss, S. Weiss, and B. Weiss, Am. J. Gastroenterology 40: 528 (1963).
21. M. L. Riccitelli and J. Weiss, J. Am. Ger. Soc. 8: 469 (1965).
22. D. Y. Graham, J. L. Smith, and D. J. Patterson, Am. J. Gastroenterology
78: 257 (1983).
190 Berger

23. W. Noll, Chemie und Technologie der Silicone, Verlag Chemie, Weinheim,
1968.
24. A. J. Barry, J. Appl. Phys. 17: 1020 (1946).
25. W. A. Piccoli and F. 0. Stark, J. Am. Chem. Soc. 77: 5017 (1955).
26. A. Weissler, J. Am. Chem. Soc. 71: 93 (1949).
27. R.R. LeVier, L'actualite chimique, 89 (1986).
28. W. D. Harkins, J. Chem. Phys. 9: 552 (1941).
29. J. V. Robinson and W.W. Woods, J. Soc. Chem. Ind. 67: 361 (1948).
30. S. Ross, Rensselaer Polytech. Inst. Eng. Sci. Ser. 63: I (1950).
31. G. Koerner, "Goldschmidt informiert ... " 3/70, 24 (1970).
32. R. D. Kulkarni, E. D. Goddard, and B. Kanner, J. Colloid Interface Sci.
59: 468 (1977).
33. S. Ross and G. Nishioka, J. Colloid Interface Sci. 65: 216 (1978).
34. This book, Chapter l.
35. K. Haubennestel, DEOS 3245482 (1982).
36. E. Pirson and J. Schmidlkofer, DEOS 2720512 (1978).
37. I. A. Lichtmann and F. E. Woodward, US 3793223 (1974).
38. S. Nitzsche and E. Pirson, DEOS 1545185 (1971).
39. K. Klein and J. Maluzi, BP 1204383 (1967).
40. S. Ross and G. Nishioka, Emulsions, Latices, Dispersions, in (P. Becher and
M. N. Yudenfreund, eds), Marcel Dekker, New York 1978, p. 237.
4 l. P. R. Garrett, J. Colloid Interface Sci. 69: I 07 (1979).
42. A. Dippenaar, lnt. J. Miner. Process 9: 1 (1981).
43. G. C. Frye and J.C. Berg,.!. Colloid Interface Sci. 127: 222 (1989).
44. V. K. Rowe, H. S. Spencer, and S. L. Bass, J. Ind. Hyg. Toxicol. 30: 332
(1948).
45. V. K. Rowe, H. S. Spencer, and S. L. Bass, Arch. Ind. Hyg. I: 539 (1950).
46. G. D. Child, H. 0. Panguin, and W. B. Deichmann, Arch. Ind. Hyg. 3: 479
(1951).
47. Ch. Gloxhuber and G. Hecht, Arzneimittel Forschng 5: 10 (1955).
48. W. E. McDonald, G. E. Lainer, and W. B. Deichmann, AMA Arch. Ind.
Health 21: 514 (1960).
49. G. L. Kennedy, M. L. Keplinger, and J.C. Calandra, J. Toxicol. Environ.
Health 1: 909 (1960).
50. M. G. Cutler, A. J. Collings, I. S. Kitt, and M. S. Sharratt, Food Cosm.
Toxicol. 12: 443 (1974).
51. E. Largent, M. Blackstone, and J. Roth, U.S. Air Force Med. Service, 1
(1950).
52. J. C. Calandra, M. L. Keplinger, E. J. Hobbs, and J. L. Taylor, Polymer
Preprints 17: 1 ( 1976).
53. R. Dailey, FDA Report No. FDA/BF-79/29 (1978).
54. FDA 21 CFR 174.340 of the United States Food and Drug Administration.
55. NF XVII; Official Monograph pub!. by the United States Pharmacopeial Con-
vention Inc. (1990).
56. USP XXII; Official Monograph pub!. by the United States Pharmacopeial
Convention Inc. (1990).
Antifoams in Pharmaceuticals 191

57. General Test and Assays, U.S. Pharmacopeia XXU, p. 1619, (1990).
58. General Test and Assays, U.S. Pharmacopcia XXII, p. 1609, (1990).
59. General Test and Assays, U.S. Pharmacopeia XXII, p. 1609, (1990).
60. General Test and Assays, U.S. Pharmacopeia XXII, p. 1523, (1990).
61. General Test and Assays, U.S. Pharmacopeia XXII, p. 1520, (1990).
62. General Test and Assays, U.S. Pharmacopeia XXII, p. 1524, (1990).
63. octylphenol ethoxylated.
64. A specification for the registration of Simethicone in Germany is in preparation.
65. J. E. F. Reynold, Martindale-The Extra Pharmacopoeia, 28th ed. Phar-
maceutical Press, London, 1982.
66. DAB = Deutsches Arzneimittel Buch.
67. "Rote Liste 1989", editor: Bundesverband der Phamiaceutischen lndustrie
e. V., Frankfurt, 1989.
68. Smith Kline & French Lab GB 1129260, (1968).
69. 0. Honecker, DEOS 2040425, (1972).
70. W. Bilhuber and P. Rados, DEOS 2517585, (1975).
71. J. A. Rider, USP 4198390, (1980).
72. J. A. Rousseau, EPA 0068566, (1983).
73. M. Wischniewski, H. Pfanz, and S. Funke, GB 1154256, (1969).
74. V. B. Surpuriya and C. A. Colleen, DEOS 2940905, (1980).
75. FDA 21 CFR 331 of the United States Foods and Drugs Administration.
76. J. Schnekenburger, Arzneimittel Forsch. 24: 142 (1974).
77. G. D. Kerr, J. lnt. Med. Res. 2 (Suppl (2)): l (1974).
78. C. van Dop, G. M. Overliet, and H. M. Smits, Pharma. Weekblad 11 I: 1093
(1976).
79. R. A. Yokel, Am. J. Hosp. Pharm. 34: 200 (1977).
80. A. Singh and H. C. Mital, Phann. Acta Helv. 52: 319 (1977).
81. P. Hagemann and K. Gamper, Schweiz. Rundschau Med. 69: 847 (1980).
82. D. Drake and D. Hollander, Ann. Intern. Med. 94: 215 (1981).
83. M. Stowasser, Z. Allgem. Med. 57: 1262 (1981).
84. K. Thoma and H. Lieb, Pharm. Zeitung 129: 121 (1984).
85. M. E. MacCara, F. J. Nugent, and J. B. Gamer, Can. Med. Assoc. J. 132:
523 (1985).
86. FDA 21 CFR. 178.3400.
87. General Test and Assays, U.S. Pharmacopeia XXU, p. 1479, (1990).
88. A. M. Rezak, J. Pharm. Sci. 55: 538 (1966).
89. H. L. M. Cox and C. J. Nijland Pharma. Weekblad 111: 973 (1976).
90. J.E. Carless, J.B. Stenlake, and W. B. Williams, J. Pharm. Pharmac. 25:
849 (1973).
91. R. N. D. Birtley et al. J. Phann. Pharmac. 25: 859 (1973).
92. H. F. Fink et al., EPA 097867, (1982).
93. P.A. Morgan, EPA 0091802, (1983).
94. H. Reuter, W. Seiter, DEOS 3436194, (1986).
95. P. W. Appel et al., EPA 0256833, (1988).
96. F. Mittelstaedt and E. Grabener, Akt. Gastrologie 6: 563 (1977).
192 Berger

97. J. A. Stead, R. A. Wilkins, and J. J. Ashford, J. Pharm. Pharmac. 30: 350


(1978).
98. K. D. Loebel, Artzliche Praxis 28: 1194 (1976).
99. R. Gladisch, R. Elfner, B. Massmer, and H. Ulrich, Ultraschall 6: 114 (1985).
JOO. P. Brockmann, Therapiewoche 31: 5882 (1981).
101. B. Fixa, 0. Komarkova, and R. Krizek Ther. Gegenw. ll J: 217 (l 972).
102. Th. Konstantinidis, Akt. Gastrologie 5: 447 (1976).
103. H. Gotzes, Therapiewoche 26: 6177 (1976).
104. J. Vivat, Acta Therapeutica 8: 79 (1982).
105. A. Frilhsorger and E. Fuchs, Therapiewoche 36: 3737 (1986).
106. H. Eveld, Med. Mschr. 31: 424 (1977).
107. J.M. Auld, Curr. Therap. Res. 26: 55 (1979).
108. W. Schroeder, Therapiewoche 30: 4945 (1980).
109. G. Neumeyer and H. J. Rogozinski, Z. Alig. Med. 55: 539 (1979).
llO. R. Virkki, P. Makela, and M. Koramo, Europ. J. Radiol. 1: 134 (1981).
111. P. J. Davis, K. A. Khan, and S. M. Sallis, Pharm. Acta Helv. 51: 378
(1976).
5
High-Performance Antifoams for the
Textile Dyeing Industry

GEORGE C. SAWICKI Dow Coming Europe, Brussels, Belgium

I. Introduction 193
H. Causes of Foam in Textile Dyeing 194
A. Textile dyeing auxiliaries 194
B. The dyeing process 198
III. Foam Control in Textile Dyeing 202
A. Foam stabilization 202
B. Antifoams for conventional dyeing processes 203
C. Antifoams for jet dyeing processes 211
IV. Conclusion 218
Acknowledgment 219
References 219

I. INTRODUCTION
Fortunately for the human race, simple incorporation of air in pure water or
water containing inorganic salts does not create a durable foam. It is only
through the introduction of organic surface-active agents or surfactants that
a durable foam will be generated. The stability or persistency of the foam
and the volume of the foam will depend on the nature and concentration of
the surfactants present and the degree of air entrainment within the process.
Therefore, any process that involves the introduction of air into a surfactant-
containing aqueous system is liable to have a foam problem. Many such
processes exist in the textile industry and preventing the formation of un-
wanted foam is often critical to their efficiency.
193
194 Sawicki

Since foam can only occur if air is incorporated into an aqueous, surfac-
tant-containing bath, the first step in tackling a foam problem should be a
complete review of the process to minimize turbulence and air entrainment,
for example, extending feed pipes to below the liquor level in tanks and
kettles, checking that agitators are well immersed in the liquid, checking
pumps to ensure that air is not being fed into the flow of liquor, and re-
moving air from the textile material prior to processing. The second step
should be to review all the surfactant-containing ingredients used in the pro-
cess, for example, reducing the concentration of each of these ingredients
to an absolute minimum, identifying alternative products that contain sur-
factants which produce less foam or a less stable foam, and minimizing the
carryover of ingredients from one process step to the next. By taking the
necessary corrective actions the foam problem should have been minimized,
but if the foam level is still unacceptable then the addition of the appropriate
antifoam will be necessary.

II. CAUSES OF FOAM IN TEXTILE DYEING


A. Textile Dyeing Auxiliaries
The largest use of antifoams in the textile industry is during the dyeing of
fibers and fabric. The medium used is predominantly water, because of its
suitability, availability, and low cost, and the process itself can be described
as the transfer of water-soluble or water-dispersible dyes from the aqueous
medium onto and into a given fiber. To obtain uniform or even dyeing there
must be unimpeded transfer of the dye between the medium and the fiber.
This can only be achieved if foam is eliminated from the dye bath, thus
ensuring that the surface of the fiber is in direct contact with the dyeing
medium throughout the dyeing process. In many cases this requirement can
only be met through the addition of the appropriate antifoam.
The choice of dye will very much depend on the required color, the nature
of the fiber being dyed, and the dyeing process selected. One feature that
all the dyes have in common is that they are all water soluble at some stage
during the dyeing process. While the dyes are present in the dye bath in
their water-soluble state, they can be classified as surface-active materials
since they consist of a hydrophobic color component and hydrophilic, water-
solubilizing groups. As surface-active species they are able to promote and
stabilize foam. However, in general, the foam that can be generated by the
dye molecules themselves is neither copious nor stable.
The main categories of dyes are acid dyes, basic dyes, disperse dyes,
reactive dyes, direct dyes, vat dyes, and sulfur dyes. Normally, the nature
of the fiber dictates the category of dye that is chosen. Acid dyes are used
Antifoams for Textile Dyeing 195

primarily to dye fibers with cationic groups such as nylon, wool, silk, and
modified acrylics. Basic dyes are used to dye fibers with anionic groups
such as acrylics and modified polyesters. Disperse dyes are used to dye
hydrophobic fibers such as polyesters and nylon. Reactive dyes are used to
dye fibers such as cotton, nylon, wool, and silk that contain chemical groups
with which they can react, and direct dyes, vat dyes, and sulfur dyes are
used to dye cotton. However, simply selecting the appropriate dye for a
given fiber will not ensure level or even dyeing. In order to achieve this
aim, the dye bath must also contain the appropriate level of various other
ingredients, commonly referred to as textile-dyeing auxiliaries.
The first example is the use of leveling agents with acid dyes. Acid dyes
are sodium salts of dye molecules which contain one or more sulfonic acid
groups, the presence of the sulfonic acid groups providing the necessary
solubility in aqueous media. They are used extensively in dyeing fibers such
as nylon which contain potential cationic groups. In the case of nylon, these
dyes are applied under acidic conditions, so that the amine dye adsorption
sites present on the fiber are protonated, thereby making bonding possible
through complex formation between the cationic dye adsorption sites and
the anionic dye molecules [l]. In practice, mixtures of dyes are more com-
monly used than single dyes to produce the required shade. This is an added
complication, because the dyes with the higher affinity for the limited num-
ber of amine groups on nylon will be adsorbed preferentially, and because
they are more firmly bonded tend not to level by migration effectively. This
also leads to highlighting of irregularities in the yam introduced by irregular
tension during weaving or nonunifonn heat setting [2]. Anionic leveling agents,
such as colorless sulfonic acids [2], are usually added to the dye bath to
improve the leveling and migration properties of the dyes. These materials
preferentially adsorb on the dye receptor sites and temporarily block them,
slow down the rate of exhaustion of the dye and by so doing help achieve
more level dyeing [ 1]. An alternative method is to complex the dye with a
cationic surfactant and thereby reduce the rate of adsorption of the acid dye
on the fiber. When using this approach, nonionic surfactants must also be
added to maintain the dispersion of the complex in the dye bath [2]. The
presence of these surface-active, leveling agents will have a significant im-
pact on the foam propensity of the dye bath.
Basic dyes are water-soluble dyes that contain a cationic dye molecule.
They are used in dyeing acrylic fibers which have structural anionic groups
present [l]. Under neutral or basic conditions, bonding occurs through com-
plex formation between the anionic dye adsorption sites of the acrylic fibers
and the cationic dye molecules. Diffusion of these dyes to these sites is
negligible below the glass transition temperature (about 65°C in water), but
is rapid at the normal dyeing temperature of 96-100°C. Typical dyeing time
196 Sawicki

is 1-2 hr but can be shortened by dyeing under pressure. Leveling of basic


dyes can be a significant problem that can often be traced back to the dif-
ferent rates of exhaustion of the individual components of the particular dye
shade. To overcome the poor leveling, colorless, cationic surfactants are
usually added to the dye bath. They preferentially adsorb on the dye receptor
sites, slow down the rate of adsorption and this in tum leads to improved
uniformity of color [l]. An alternative method is to add anionic surfactants
which can form complexes with the dye. This reduces the amount of cationic
dye available for adsorption and hence reduces the rate of adsorption of the
dye on the fabric. To achieve adequate dispersion of the anionic surfactant-
dye complex in the dye bath, nonionic surfactants must also be added [2].
As with acid dye leveling agents, these additional surface-active materials
will have a significant effect on the foam propensity of the dye bath.
Unless chemically modified, hydrophobic fibers such as polyester lack
reactive dye sites and are dyed almost exclusively with disperse dyes. These
dyes have a very low solubility in water and have to be added to the dye
bath as very fine, uniform, stable dispersion. To achieve the correct level
of dispersion, the dyes are wet-milled in the presence of high levels of sur-
face-active, dispersing agents such as lignin sulfonates to give a dye in the
liquid or paste form, or subsequently spray-dried to give the powder form
[3]. This implies that every time a disperse dye is added to the dye bath, a
significant quantity of surface-active, dye dispersants will also be added. In
some cases, additional dye dispersant may need to be added to the dye bath
to ensure that the disperse dye remains properly dispersed. This is most
likely when dyeing pale-to-medium shades, when the quantity of dye dis-
persant added via the dye is insufficient to give the required concentration
in the dye bath [4]. Dispersing agents based on naphthalene sulfonic acid-
formaldehyde condensation products and especially cresol sulfonic acid-
formaldehyde condensation products have been shown to be particularly ef-
fective, and even at high concentrations show little or no retarding [5].
The levelness of polyester dyeing is not only dependent on the stability
of the disperse dyes in solution, but also on the affinity of the fiber for the
dyes and the ability of the dyes to migrate in the fiber at the dyeing tem-
perature. By adding leveling agents such as ethoxylated fatty acids and
ethoxylated nonyl or tributyl phenols [5], it is possible to increase the sol-
ubility of the disperse dyes which in tum leads to improved dye migration
and increased dye penetration. In general, anionic surfactants are used to
improve the dispersion stability of the dye and nonionic surfactants are used
as leveling agents. Although the latter improve leveling, they are not effec-
tive enough to cover highlighting of irregularities in the yam introduced by
irregular tension during weaving or nonuniform heat setting.
Antifoams for Textile Dyeing 197

At temperatures below 100°C the rate of dyeing of polyester with disperse


dyes is very slow. To accelerate the process, dye carriers or accelerants are
added to the dye bath. These include materials such as o-hydroxydiphenyl,
methyl salicylate, tripropyl phosphate, butyl benzoate, and trichlorobenzene
[6]. Their addition can increase the dyeing rate by as much as 50 times that
achievable in their absence. These, essentially water-insoluble, organic com-
pounds, swell the fiber and allow the molecularly dispersed dye to diffuse
into the fiber more rapidly [6]. As well as increasing the rate of diffusion
into the fiber, they also strip excess dye from the fiber. The dyeing process
is complete when the rate of uptake of dye equals its rate of removal and
wash fast, level dyeing is achieved. To overcome the slow dyeing rate, high-
temperature and pressure processes have been developed. In these processes,
much lower levels of dye carrier are required since the rate of diffusion of
the dye both into and out of the fiber occurs far more rapidly at these ele-
vated temperatures. Legislative pressure is hastening the trend toward higher
temperature and pressure since very few of the carriers can be regarded as
environmentally desirable. Dye carriers are normally supplied as emulsions
to achieve the correct level of dispersion in the dye bath and prevent the
essentially insoluble, active material from separating out and causing spot-
ting or uneven dyeing. In the case of solid, carrier-active compounds, sol-
vents are used to solubilize the materials and enhance the activity of the
finished product. This means that whenever a dye carrier is introduced into
a dye bath, surface-active emulsifiers such as sodium dodecylsulfate [2] used
in formulating the emulsion will also be added.
Clearly the composition of the dye bath is more than just a mixture of
fabric, dye, and water, and will necessarily include other additives such as
dyeing auxiliaries. This implies that the dye bath will contain a number of
different surface-active materials, and it is these materials that promote and
stabilize foam in various dyeing processes. The surface-active materials will
include the actual dyes themselves, when they are present in their water-
soluble state, but more importantly the surface-active components of the
dyeing auxiliaries. Some examples already given include anionic, nonionic,
and cationic leveling agents, dye dispersants in the formulation of disperse
dyes and emulsifiers used in formulating dye carriers. In addition there will
also be other auxiliaries such as wetting agents to ensure that the fiber or
fabric comes into direct contact with the dyeing medium, and fiber lubricants
to reduce chafe marks.
A description of the composition of a dye bath would not be complete
without at least some indication as to the concentrations of the various dyeing
auxiliaries that are likely to be present. Taking one of the major dyeing
processes, namely the dyeing of polyester with a disperse dye, a typical dye
bath could contain [5]
198 Sawicki

0.5-5% by weight of substrate of a disperse dye


0.3-1 g/L leveling agent
0.5-2 g/L dispersing agent
0.0-1 g/L dye carrier
This means that significant quantities of surface-active materials are inten-
tionally added to achieve level or even dyeing.
Finally, there is one further source that is often neglected and that is
accidental carryover of detergent surfactants from the previous scouring step,
usually resulting from inadequate rinsing prior to the dyeing step. This means
that virtually every dye bath is the possible source of a significant foam
problem.

B. The Dyeing Process


Within the dyeing machine, there must be sufficient movement of the liquor
to ensure uniform penetration of the fiber assemblies, uniform temperature
throughout the liquor, and rapid dilution of the concentrated dyestuff and
auxiliaries before they come into contact with the fiber assemblies. There
are three main ways of achieving these goals [7].
1. Circulating the Dye Liquor through Stationary
Substrate
In package dyeing machines, yarn is wound around a stainless steel "for-
mer" to give a "cheese." These cheeses are loaded onto perforated steel
tubes mounted in the dyeing vessel, and dye liquor is then pumped through
the perforated steel tubes, the formers, and the fabric before flowing back
to the pump for the process to continue. In beam dyeing, cloth is wound
onto a horizontal, hollow, perforated cylinder and dye liquor is pumped
radially through the beam. To ensure level dyeing, package and beam dyeing
machines have an automatic reverse switch which changes the flow direction
of the liquor on a preset schedule. In loose stock dyeing machines, the fibers
are packed into a vessel between two perforated plates. The liquor is then
circulated through the bottom plate, through the mass of fibers and out through
the top plate, before returning to the pump. Since level dyeing is not so
critical with loose stock dyeing, machines tend only to pump the dye liquor
in one direction.
2. Movement of Substrate without Mechanical
Movement of Liquor
This technique is used in chain warp dyeing, jig dyeing, beck dyeing, and
continuous dyeing. In chain warp dyeing, several warp ropes are pulled through
a series of vats of dye liquor until the correct shade has been obtained. In
Antifoams for Textile Dyeing 199

jig dyeing several hundreds of metres of cloth arc wound on a beam roller,
and then transferred onto another roller via a dye bath. Once the cloth has
been wound onto the opposite beam roller, the direction is reversed and the
cloth is rewound onto the original roller. This is continued until the cloth
is dyed to the correct shade. In beck or winch dyeing, an elliptical roller
lifts the fabric from the vertically faced side of the dye filled beck and trans-
fers it to the gradually sloping side of the machine. The fabric slides down
along the slope before once again being lifted out by the roller. The elliptical
roller arranges the fabric in folds before it falls back into the dye liquor.
The fabric is circulated through the dye bath in this manner until the correct
shade is obtained. The modern beck is constructed of stainless steel to be
operated at atmospheric pressure or if constructed as a heavy-walled, scal-
able unit it may be used at high temperatures and pressures. In continuous
dyeing, the dye is pad applied to the fabric. The dye is applied by passing
the fabric through the dye bath, excess dye is then removed by passing it
between a pair of squeeze rollers before going on to be processed to achieve
adequate fixation of the dye. This technique is widely used in dyeing long
lengths of nonstretch, woven goods.
3. Movement of Substrate and Circulation of Liquor
This procedure is used in jet dyeing. In this batch process the circulation of
the liquor is actually used to move the fabric [8,9). The fabric, which is in
rope form, is transported by the movement of liquor through a venturi jet
(Fig. 1). The cloth travels vertically through the center of the jct, and at the
throat of the jet is met by a circumferential, converging jct of dye liquor
which emerges from a ring-shaped slot completely surrounding the fabric
(Fig. 2). The fluid simultaneously penetrates and propels the fabric and pro-
duces tremendous interchange between the fluid and the fabric. The fabric
emerges from the cloth guide tube where it is met by a doffing jet which
changes the direction of motion of the material and helps pile the material
in loose folds in the cloth storage chamber, so no crushing or rope marking
can occur. Here the fabric is completely submerged, but the high-volume
flow through the chamber ensures the mass of fabric is kept moving. As the
cloth emerges from the chamber it passes through metering rolls which both
control the cloth speed and reduce the tension on the fabric, as now it only
has its own weight to bear to the rolls and not the full height to the jet. A
centrifugal pump recirculates the entire dye bath through the jet every 20-
30s. Operating temperatures up to 160°C are combined with the high fluid
energy and very short dye times result. Modern jet dyeing machines are
fairly gentle to the fabric, while providing the required intimate contact be-
tween the dye and the fabric necessary for uniform dyeing [9]. The big
advantage that this technique has over beck dyeing is the energy saving that
JET
Q~
/l):u . . ·. ·. • • rn . . .· '· ' u-- CLOTH GUIDE TUBE 8
UNLOADING
REEL -

THROTTLE VALVE
~ ~
.
I EIII DOFflNG JET

-
METERING ROLLS

I
LOADING PORT

HEAT EXCHANGER

){~ - -.--- ... >iii.L ___ -~c


i __
I I
5
\ : CLOTH SlORAGE
ADD DYE TANK ~ .~ CHAMBER

CENmlFUGAL ~------ DRAIN


PUMP II I I II ....._

ADD DYE PUMP~

fJ)
FIG. 1 Schematic cross section of a Gaston County Jet Dyeing Machine. (From Ref. 8. Reproduced with kind permission of ;c;
Gaston County Dyeing Machine Co., North Carolina.)
!!5:
Antifoams for Textile Dyeing 201

FIG. 2 Schematic cross section of the main jet in a Gaston County Jet Dyeing
Machine. (From Ref. 8. Reproduced with kind permission of Gaston County Dyeing
Machine Co., North Carolina.)

results from the increased cloth-to-liquor ratio, especially when using high-
temperature and -pressure dyeing conditions. Typical cloth-to-liquor ratios
would be of the order of 1: 10 for a full load. Under certain conditions,
cloth-to-liquor ratios as high as 1 : 5 can be used. The high circulation rate
ensures the homogeneity of the dye bath from a chemical and temperature
standpoint and leads to more level dyeing than with other dyeing techniques
such as jig or winch dyeing [9].
The Gaston County jet dyeing machine described above is one of many
different designs available. What is common to all jet dyeing machines is
the use of high-velocity circulating liquor to transport the fabric rope through
the machine. The high velocity is achieved either via a large circulating
202 Sawicki

pump and venturi, as in the Gaston County machine, or via an overflow


system. In the latter, the venturi is replaced by a system in which liquor is
pumped from the lower half of the machine to a reservoir in the upper half.
Liquor then flows down a narrow tubular sluice under gravity carrying the
fabric with it. Again a driven winch is used to assist and control fabric
circulation [9].
The jet dyeing machines described above are known as partially flooded
machines since the dye liquor only fills part of the total volume. Due to the
high level of liquor turbulence, caused by the rapid rate of fabric and liquor
circulation in the jet dyeing machine, such machines represent a major foam
control problem. If foam is allowed to form, the pump may well start cav-
itating. This in tum will lead to loss of pump efficiency, and tangling and
stoppage of the fabric rope. In order to continue the dyeing process, the
dyeing machine must be cooled to below the boil, the tangle located and
dealt with. The liquor must then be reheated to the dyeing temperature. This
is both costly and time consuming. In order to eliminate the foam problem
fully flooded machines were introduced. In these machines the total free
volume of the machine is filled with the dye liquor. As a result no air can
be entrained into the circulating liquor and hence no foam will be formed.
Such a development also allowed the use of dyes and dyeing auxiliaries that
could not be used in partially flooded machines because of their high ten-
dency to promote foam.
Unfortunately, because the fabric is in constant motion, tangling is a fre-
quent occurrence [9]. This coupled with a need to reduce energy consump-
tion has led to a move back to partially flooded machines in which lower
liquor volumes are required.
In every one of the partially flooded processes described, there is suffi-
cient turbulence to entrain air in the dyeing medium. Since the dyeing me-
dium already contains a significant number of surface-active materials, foam
formation during the dyeing process is inevitable. The volume of foam and
its persistence will depend on the nature and the concentration of the sur-
factants present, and the degree of turbulence within the process. Persistent
foam will cause uneven dye adsorption, fabric flotation, pump cavitation,
fabric tangling, wasted volume in process equipment, and general house-
keeping difficulties. It is therefore not surprising that the largest use of an-
tifoams in the textile industry is in the textile dyeing area.

llL FOAM CONTROL IN TEXTILE DYEING


A. Foam Stabilization
A foam consists of gas entrapped within thin films or lamellae of a liquid.
Foaming is always accompanied by an increase in the interfacial surface area
Antifoams for Textile Dyeing 203

of the system and hence its total free energy. Since systems always seek to
reduce free energy, foams must be thermodynamically unstable. Foam for-
mation requires a net input of energy and it is the mechanical agitation pres-
ent in the process that provides the energy needed to entrain the gas initially.
The ease of foam formation in an aqueous medium is determined in part by
the requirement that stabilizing surface tension gradients exist. The surface-
active materials present in the continuous phase can adsorb at newly created
surfaces to create such gradients thereby stabilizing the gas that has been
entrained and hence the foam that has been produced.
Depending on the conditions prevailing, the stability of the foam gen-
erated can vary from milliseconds to almost unlimited duration. The pres-
ence of materials in the dye bath that can adsorb at the freshly created in-
terface implies that foam formation is inevitable when air is entrained in the
dye liquor. However, the stability of the foam generated is dependent on a
number of other factors. These include
Surface elasticity
Surface viscosity
Bulk viscosity
Electrical double layer and entropic repulsion
Gas diffusion
Evaporation
The relative importance of each of these will depend on the materials that
arc present in the dye bath. Unfortunately, it is only possible to guess at the
order since these factors have not been investigated for dye bath composi-
tions. A more comprehensive coverage of the theory of foam stabilization
is given in Chapter 1.

B. Antifoams for Conventional Dyeing Processes


In many practical applications the foam lasts long enough to interfere with
the physical and chemical processes being carried out, and steps must be
taken to prevent it occurring. It is this need that has led to the development
of a vast number of antifoaming agents whose primary role is to destabilize
the foam as it is being generated through air entrainment.
Though antifoams are primarily judged by the persistence of their anti-
foaming action in a given foaming medium, this is by no means the only
criteria for choice. Other equally important attributes include
Must eliminate existing foam as well as prevent further foam from forming
Must be storage stable
Must be chemically stable under dyeing conditions
Must be easy to disperse in the dye bath
204 Sawicki

Lack odor
Lack color
Must not react or interfere with dyes or auxiliaries present in the bath
Must not deposit on fabric and machinery causing spotting or staining of
fabric
Must be safe to both humans and the environment
Such conditions can often be met by proper formulation of the product.
However, no one product is ever likely to give the optimum performance
in all the possible dye bath conditions that could be met. Ideally, to avoid
any subsequent problems, one should test the compatibility of the antifoam
with all the other dye bath components under the conditions that will be
encountered within the process prior to use in production. Such preliminary
testing will avoid the risk of instability in the dye bath leading to imperfect
dyeing.
1. Typical Antifoams Used in Textile Dyeing
In textile processing a number of different categories of antifoaming agents
are used. These include formulated products based on fatty acids, fatty al-
cohols [10], and their ethoxylates, polyoxyethylene and polyoxypropylene
block copolymers [11], dispersions of mineral oils or silicone fluids and
hydrophobic silica [ 12], silicone glycols [ 13], and phosphoric acid esters
[14-16] among others. The choice is often governed by the specific surface-
active agents present in the bath and the conditions that will be encountered
in the dye bath.
Unfortunately, each class of foam control agent has inherent deficiencies
[ 17]. Alkali earth metal or aluminum salts of fatty acids can leave undesir-
able deposits on the processing equipment which in extreme cases can cause
pipes and jets to clog. Aliphatic alcohols have relatively poor foam control
performance and in addition have an unpleasant odor that can cause nausea
to workers, even at low concentrations in the atmosphere surrounding the
textile bath. Polyoxyethylene and polyoxypropylene block copolymers have
relatively poor foam control performance which significantly limits their use.
Emulsions of hydrophobic oils and silica are particularly effective as foam
control agents; however, they are susceptible to destabilization when used
in processes where the dye liquor is subjected to high temperatures, above
100°C, and high shear. If the emulsions break in use, the hydrophobic oil
deposits on the textile resulting in undesirable spotting or staining of the
goods being dyed, and simultaneously the desired foam control action is
lost. This problem can be overcome by using silicone glycols, rather than
hydrophobic oils and silica, but such materials have much poorer foam con-
trol ability. Phosphoric acid esters also have relatively poor foam control
performance which means that their use is limited to selected dyeing processes.
Antifoams for Textile Dyeing 205

2. Improving Antifoams through Formulation


The challenge to the textile auxiliary formulator is to provide the textile dyer
with a foam control agent that can be used in a wide variety of dyeing
processes without any negative impact on either the dyeing process or the
goods being dyed. As yet there is no universal solution, however, significant
progress has been made over the last two decades through better formulations.
The first example shows that through proper formulation it is possible to
overcome some of the drawbacks associated with using the aluminum salt
of a fatty acid as the foam control agent [18], by combining the fatty acid
salt with surfactants, aliphatic alcohols, and hydrophobic mineral oils. A
typical formulation would be
4% Aluminum distearate
3% Phosphoric acid esterified with polyoxyethylated p-nonyl phenol
6% 2-ethyl-n-hexanol
87% Paraffin oil
by weight. Unlike previous formulations containing aluminum stearates such
formulations are very stable and will control the foam when dyeing wool
with 1: 1 and 1: 2 metallic, acid, or reactive dyes, dyeing nylon with acid
or disperse dyes, dyeing polyester with disperse dyes, dyeing cotton with
reactive and direct dyes, and acrylics with cationic dyes. An alternative to
using aliphatic alcohols is to use a propoxylated, branched, aliphatic acid,
such as 1,2-propylene glycol mononeodecanoate [ 17]. Though the foam con-
trol performance is similar to that of isooctanol, it has a much lower odor.
Typical addition levels that would be required are from 0.2% to 0.5% by
weight based on the weight of the textile-dyeing liquor. Although it can be
used in combination with virtually all dyes, it is particularly effective when
used in combination with acid dyes in either batch or continuous dyeing
processes.
To overcome the deficiencies of polyoxyethylene-polyoxypropylene block
copolymers, it has been proposed to combine them with a silicone antifoam
compound and use a silicone glycol copolymer as the dispersing agent [19].
By using this composition it is possible to obtain a stable, highly effective
foam control agent. A typical composition would be
52% Polypropylene glycol with a molecular weight of 4100
40% Silicone antifoam compound consisting of polydimethylsiloxane fluid,
silicone resin, and silica
8% Silicone glycol copolymer as dispersing aid by weight
The formulation of the silicone glycol copolymer being
Me 3SiO(Me 2Si0)64(MeSi0) 6 SiMe 3

I
(C:ll2)3(0C:3ll6)100C:4ll9
206 Sawicki

The dispersing aid maintains the dispersion of the silicone antifoam com-
pound in the polypropylene glycol.

3. Formulating Antifoams into Textile Dyeing Auxiliaries


Rather than adding the foam control agent separately to the dye bath, it is
becoming more common to formulate the antifoam into the textile-dyeing
auxiliary to give a low-foaming auxiliary. The formulation of such an aux-
iliary could be [ 1,20]

15-30% Nonionic, anionic, or cationic textile-dyeing auxiliary


10-50% Fatty acid esterified with a polyoxyethylene-polyoxypropylene block
copolymer
55-70% Water or water-soluble solvent

by weight. The typical formulation of the foam control agent [ 1] would be


the esterification product of 2 moles of tallow acid and a polyoxyethylene-
polyoxypropylene block polymer having a molecular weight of 2600. Pref-
erably, the fatty acid should contain at least 30% by weight of oleic acid in
order to improve the solubility of the ester in the formulation to give a clear,
stable, homogeneous mixture. The textile-dyeing auxiliary is a leveling or
wetting agent normally added to ensure even and level dyeing. Typical non-
ionic dyeing auxiliaries that could be used in the above formulation arc poly-
oxyethylated and/or polyoxypropylated saturated fatty acids, fatty alcohols,
fatty amines, fatty amides, alkyl phenols, and glycols. Typical anionic aux-
iliaries are the sulfated or sulfonated versions of the nonionic surfactants
already described, alkylbenzene sulfonates and alkyl sulfonates. Typical ca-
tionic auxiliaries are salts of the alkyl pyridinium ion, methylalkyl piperi-
dinium ion, and tetraalkyl ammonium ion. The resultant mixture when added
to the dye bath gives level dyeing, and the dye bath exhibits practically no
foaming. However, these formulations can only be used in dyeing processes
where the pH remains between 2.5 and 8.0 during dyeing. Most importantly,
they cannot be used in alkaline dye baths because the fatty acid ester foam
control agent becomes saponified and hence inactivated. This presents a lim-
itation as regards the dyeing processes in which they can be used.
When dyeing synthetic fibers with insoluble or sparingly soluble dyes,
uneven dyeing with poor fastness to rubbing can often occur due to inade-
quate dispersion of the dyes in the dye liquor, for example, in high-tem-
perature dyeing of polyester fibers with disperse dyes. In order to overcome
this problem, dispersing agents are generally added in conjunction with lev-
eling agents. However, these auxiliaries normally have only either dispers-
ing or leveling properties, and it is necessary to use mixtures of both to
achieve the best results. It has been found that alkali metal or ammonium
salts of the addition products of polypropylene glycol and polyhydric ali-
Antifoams for Textile Dyeing 207

phatic alcohols, such as glycerol, reacted with sulfamic acid not only have
the ability to stabilize the dye dispersion, so that neither aggregation nor
precipitation of the dyestuff occurs during the dyeing process, but also have
a low-foaming tendency [21]. This leads to level dyeing which is fast to
rubbing.
However, in order to achieve even lower foaming, it is recommended to
incorporate adipic acid di-2-ethylhexyl ester and polydimethylsiloxane as
antifoaming agents. The amount of antifoaming agent used being 15% to
60% by weight based on the weight of the auxiliary. Water-miscible alcohols
are also added to improve the dispersibility of the formulation in the dye
bath.
Using a similar approach, it is possible to obtain a low-foaming, wetting
agent by formulating together a wetting agent and a foam control agent [22-
24]. A preferred formulation [221 is
10-20% Anionic surfactant
15-35% Insoluble or sparingly soluble propoxylate
0.1-2% Silicone antifoam compound
40-75% Water
by weight. The anionic surfactant could be for example

where X is hydrogen, ammonium, or an alkali metal ion. Such surfactants


are well established as wetting and leveling agents in textile-dyeing pro-
cesses. However, when they are used by themselves, they produce copious
quantities of undesirable foam. The insoluble or sparingly soluble propoxy-
late referred to above is either the reaction product of I mole of a polyhydric
alcohol (such as ethylene glycol, polypropylene glycol, glycerol, pentaer-
ythritol, sorbitol, trimethylolethane, or trimethylolpropane) with 30 to 120
moles of 1,2-propylcne oxide or the reaction product of a monoalkylamine,
monoalkylolamine, or polyalkylenepolyamine (such as monoisopropanolam-
ine or ethylene diamine) and 1,2-propylene oxide. When used in conjunction
with the anionic surfactant, the foaming tendency is lower than when the
surfactant is used on its own. The silicone antifoam compound provides an
additional means of reducing the foam generated by the anionic surfactant
and this helps to enhance the wetting of the textile goods by the wetting
agent. Optionally, a water-insoluble, nonionic surfactant may also be in-
cluded, such as the reaction product of 2-ethyl-n-hexanol and ethylene oxide
or p-nonyl phenol and 3 moles of ethylene oxide among others [22].
Such compositions can be used in both alkaline and acid baths in the pH
range from 1 to 12 and over a wide temperature range from 20°C to 120°C.
They can be successfully used in a wide variety of processes, such as dyeing
208 Sawicki

wool with 1: 1 or 1: 2 metallic, acid, or reactive dyes, exhaustion and con-


tinuous dyeing processes for nylon fibers with acid or disperse dyes, dyeing
polyester fibers with disperse dyes, dyeing cellulosic fibers with reactive
and direct exhausting dyes, and dyeing acrylic fibers with cationic dyes.
4. Antifoams Based on Water-Insoluble Oils and Silica
Of the different types of foam control agents described earlier, the most
effective are those based on dispersions of hydrophobic silica in water-in-
soluble oils. With relatively mild surfactants and low-shear conditions, the
most cost-effective water-insoluble oils are vegetable oils, or more com-
monly, mineral oils such as paraffin oils, naphthenic oils, kerosene, and
similar petroleum fractions [12]. A spreading agent is often incorporated to
improve dispersion throughout the foam system. When appropriately for-
mulated, they are very effective at knocking down existing foam [25].
With the low surface tension and high agitation encountered in many tex-
tile dye baths, more effective foam control is achieved by using polydi-
methylsiloxane fluids as the water-insoluble oil. They are most often used
where prevention of foam for the duration of the process is required and
either high levels of agitation and/or high levels of low-surface-tension sur-
factants are present. They consist of proprietary combinations of polydi-·
methylsiloxane polymers or silicone fluids:
Me 3SiO(Me2SiO)nSiMe 3
and finely divided, hydrophobic silica, and are more commonly known as
silicone antifoam compounds. The fluids are colorless, essentially odorless,
chemically stable, unreactive materials. They are insoluble in water, difficult
to emulsify, have a very low surface tension, typically 21 mN m- 1 at 25°C,
and spread spontaneously over the surface of most surfactant-containing so-
lutions. In addition, they have been shown to be safe to both humans and
the environment. However, on their own they show little or no foam-inhib-
iting ability in aqueous foaming media. It is only when these fluids are com-
bined with finely divided, hydrophobic silica particles that an effective foam
inhibitor is produced [12]. An example of a formulated silicone antifoam
compound is given below [26]:
88% Polydimethylsiloxane fluid having a viscosity of 350 cSt at 25°C
9% Siloxane resin composed of (CH 3)JSi00 .5 and Si02 units in which the
ratio of (CH 3)JSi00 5 units to Si0 2 units is within the range of 0. 6: l and
1.2: 1
3% Silica aerogel
by weight. The method described to prepared the product is to mix the po-
lydimethylsiloxane fluid and silicone resin with heat at 70°C to 120°C for
Antifoams for Textile Dyeing 209

TABLE 1 Typical Formulation of a 10% Active Silicone Antifoam Emulsion

Percent by weight Ingredient

10.0 Silicone antifoam compound


4.0 Emulsifiers:
polyoxyethylene monostearate and glyceryl monostearate
2.0 Stabilizers: water-soluble thickeners
84.0 Water
As required Preservatives

1 to 3 hr. Then after adding the silica heat the resultant mixture to 130°C
to 175°C for a period of approximately 2 hr.
Even though there has been a significant amount of speculation concern-
ing the individual roles of the fluid and the silica [27-29], it is well estab-
lished that a combination of these two materials provides an antifoam that
will perform effectively in virtually all the foam control applications met in
the textile industry. The art of formulating an effective silicone antifoaming
agent is to combine the silicone fluid and silica in such a way that the most
efficient foam control agent is produced.
However, the products are rarely used as such in textile processes because
of the difficulty of dispersing such high-viscosity materials effectively in the
textile bath at very low addition levels. To achieve the required level of
dispersibility, these products are introduced into the process as emulsions
or dispersions. This also overcomes the difficulty in handling the addition
of the very small quantities of antifoam compound required to control the
foam, and the potential risk of spotting and staining problems if overdosed.
Dispersibility of the antifoaming agent in the foaming medium is important
because it affects the ability of the antifoam to perform its foam-breaking
function. If the antifoam is not adequately dispersed, it will merely sit on
the surface of the bath as an insoluble mass of liquid and no appreciable
antifoaming action will be observed. To achieve the correct level of dis-
persibility in the bath, the combination of silicone fluid and silica is supplied
as a 10% to 30% active emulsion. A formulation for a typical 10% emulsion
is given in Table 1 [30]. Textile mills in general prefer to use low-solids
emulsions as they can be used as received and thereby minimize the potential
for overuse and losses due to waste in handling. These emulsions are used
in a variety of textile-dyeing processes, as well as many textile preparatory
and finishing operations, and effluent treatment processes. In textile dye
baths they are commonly used where temperatures encountered are below
100°C. They are not recommended for processes that require high-temper-
ature dispersibility and/or shear-stable antifoam products [30].
210 Sawicki

The intrinsic foam control ability of an antifoam emulsion is directly re-


lated to the performance of the active material. However, the efficiency of
the resultant emulsion will also depend on formulating variables such as the
particle size distribution of the active material, the level of active material,
the type of emulsifiers and thickeners used in manufacturing the emulsion,
and their concentrations. It has been found that if the particle size of the
active ingredient in the emulsion is less than 2 microns, the product shows
poor antifoam efficiency [31]. On the other hand, if the particle size is of
the order of 50 microns or more, the emulsion has poor shelf stability unless
it contains very high levels of thickeners to retard separation. In addition,
large particle size emulsions will "oil out" on dilution in the foaming me-
dium resulting in oil spotting on the textile goods and machinery. In prac-
tice, the particle size is carefully controlled somewhere between these ex-
tremes through the careful selection of the emulsifiers, thickeners, and the
emulsification equipment used in their manufacture [321.
Another important consideration is the suitability of the emulsion for-
mulation for the end application. A good example is the choice of thickener
system used to retard creaming in these emulsions [32]. Some of the earliest
emulsions used methyl cellulose as the thickening agent. This is a very ef-
fective thickening agent, but unfortunately it is insoluble in water above
about 80°C. When emulsions of this type were used in textile-dyeing pro-
cesses at or near 100°C, the methyl cellulose would come out of solution
taking the antifoam active ingredient with it. This resulted in dye stains
caused by adsorption of the dye on the insoluble cellulose and subsequent
deposition on the fabric and loss of foam control efficiency due to depletion
of the active material from the foaming medium. To make matters worse
more antifoam emulsion would be added to compensate for the loss of foam
control efficiency. Modem, silicone antifoam emulsions use nongelling, cel-
lulose ethers such as hydroxy ethyl cellulose, alginate derivatives and syn-
thetic carboxy vinyl polymers as thickeners ! 32]. With improved emulsion
formulations, they have shed their poor image and are used successfully in
preparatory, finishing and dyeing processes. The levels of addition will very
much depend on the concentration and nature of the surface active ingre-
dients present in the process. Typically it would be 50-300 ppm active ma-
terial in open beck dyeing and 1-10 ppm in wastewater treatment where the
level of agitation and surfactant concentration is relatively low.
It is preferable that the antifoaming agent is added at the start of the
dyeing process, because many of the foam problems are caused by air en-
trainment in the system when circulation begins. If possible it should be
added even before the dyes and the auxiliaries [33]. However, this must
be done carefully to avoid overuse with subsequent risk of oil spotting.
Alternatively, the antifoaming agent may be added, as a very dilute dis-
persion, continuously or semicontinuously during the course of a particular
Antifoams for Textile Dyeing 211

process. This is normally the safest and most cost-effective method [32].
It must be stressed that other industries have different needs, and so not
all silicone antifoam emulsions will meet the stringent requirements en-
countered in controlling foam in textile processes.

C. Antifoams for Jet Dyeing Processes


The introduction of jet dyeing machines for rapid dyeing of polyester pre-
sented a major foam control challenge to the antifoam formulator, especially
when dyeing darker shades where higher levels of disperse dyes are re-
quired. Excessive foaming was a serious problem in production because of
the high-shear and high-temperature conditions encountered, especially in
partially flooded machines. Even by carefully selecting low-foaming, dis-
perse dyes and textile-dyeing auxiliaries the level of foam was often too
high and would interfere with the dyeing process. For example, the presence
of foam would cause the centrifugal pump to cavitate, which in turn im-
paired the action of the jet and reduced the speed of fabric circulation through
the machine. In addition this could also lead to fabric flotation, tangling,
and fiber breakage as well as uneven dyeing ll3]. Jet dyeing machines also
exposed weaknesses in the available antifoam emulsion systems. Under the
very severe conditions of high shear and high temperature encountered in
these machines, the emulsions would often break and the oil phase would
accumulate on the inside of the jet causing oil spotting on the fabric. This
gave silicone antifoam emulsions a poor image in the industry until new
formulations were developed.
1. Antifoams Based on Silicone Glycol Copolymers
The first approach was to develop more stable emulsions with smaller par-
ticle size that could better withstand the high-shear conditions present in the
jet dyeing process. Though these overcame the spotting problem, the anti-
foam efficiency was greatly reduced. The more successful approach was to
utilize the unique properties of a class of materials commonly known as
silicone glycol copolymers or silicone polyethers [ 13]. There are essentially
four structures that represent those most commonly found in the patent
literature:
Structure A
Me 3 SiO(MeGSiO)m(Me 2SiO)nSiMe 3
Structure B
GMe 2SiO(MeGSiO)m(Me2Si0) SiMe2G11

Structure C
R0 Si[(OSiMe2),,(0SiMeG)mOSiMe3]4-a
Structure D
RaSil(OSiMe2)n(OSiMeG)mOSiMe2Gl4-a
212 Sawicki

where R is typically a methyl group, a is typically l, and G is typically


represented by the following formulae:
~(C:l-l2)3(()C:2I-l4)x(()C:3l-16)y()R'

or
~(()C:2l-I4)x(()C:3l-16)y()R'

where R' is an end-capping group.


By varying the relative proportion of dimethylsiloxane, polyoxyethylene,
and polyoxypropylene in the molecule, it is possible to dramatically alter its
solubility characteristics. For example, it is possible to manufacture a poly-
mer that is soluble in cold water but not in hot water. The temperature at
which this occurs is known as the cloud point. Below its cloud point the
molecule acts as a conventional surfactant, whereas above its cloud point it
acts as a foam control agent. Furthermore, the higher the temperature of the
foaming medium the greater is the antifoaming efficiency of the silicone
glycol copolymer. This class of material became the basis of an acceptable
antifoam for the jet dyeing process. During the high-temperature dyeing stage,
the material is insoluble and acts as an effective antifoam. Whereas after
the dye liquor has been cooled, the material becomes soluble once more and
can be completely removed by subsequent rinsing. This means that by for-
mulating products based on this class of polymer it is possible to achieve
foam control without the risk of oil spotting. It is important to point out that
the temperature of the dye bath must be below the cloud point of the polymer
when it is added to achieve rapid and even dispersion of the antifoam in the
bath through solubilization. ()therwise, the optimum performance benefits
will not be achieved.
In general, the silicone glycol copolymers with the higher water solubility
and higher cloud points have lower foam control efficiency than those with
lower cloud points. 1-Iowever, as the solubility is decreased to achieve better
foam control performance, deposition of the antifoam on the dyed fabric
becomes more pronounced with the resultant risk of staining or spotting on
the fabric [34]. In addition, the antifoaming action of these cold-water-sol-
uble silicone glycol copolymers is inferior to combinations of polydimeth-
ylsiloxanes and silica. In order to overcome these problems, several de-
foaming compositions have been proposed. For example, the antifoam action
can be increased significantly by incorporating a suitable silica as in the case
of hydrophobic oils [35]. A relative comparison has been established by
determining the volume of a 10% by weight aqueous dispersion of antifoam
that is necessary to control the foam of 2.5 L of 0.01 % aqueous solution of
sodium alkyl sulfate that is being circulated through an open vessel at a rate
of 10 L/min at a temperature of 90°C. The results obtained were as follows:
Antifoams for Textile Dyeing 213

Conventional silicone antifoam compound 0.05 mL


Silicone glycol copolymer (no silica) 1.00 mL
Silicone glycol copolymer (with silica) 0.15 mL
As can be seen the silicone glycol copolymer is significantly less efficient
than the conventional silicone antifoam compound. However, through the
addition of 2% fumed silica, the performance can be dramatically improved.
The formula of the polymer used in this set of experiments was
Me3Si(OSiMe 2) 14 (0SiMeG)sOSiMe 3
where G has an average molecular weight of about 1500 and can be rep-
resented by the formula -(0CnH2n)yOR containing equal weights of oxy-
ethylene and oxypropylene units and R is a n-butyl group.
The method of preparation was to add the fumed silica to the silicone
glycol copolymer with stirring and subsequently heating the mixture at 100°C
to 120°C for 1 hr. The composition was also shown to control the foam in
a jet dyeing machine throughout the whole process, when the temperature
of the dye liquor was raised to 125°C, when dyeing was being carried out
for 8 hr, and when the pressure was built up and released. Careful exami-
nation of the dyed goods showed no color unevenness nor any traces of
silicone spotting.
Alternative structure for the silicone glycol copolymer have been pro-
posed [36]:
GMe 2Si(OSiMe2 )a(0SiMeG')b(OSiMe2 ) 0 0SiMe 2G
where
G' = -(0SiMe2) 0SiMe2G
= -(OC3H6 )iOC2H4)mOR
0

G
The foam control composition as described consists of 92.5% by weight of
the glycol, where a = 6, b = 2, n = 21, m = 6, and R = C 4H9 , and 7 .5%
by weight of a hydrophobic fumed silica with a surface area of 200 m 2 /g.
To this composition may also be added one or more nonionic emulsifiers
with HLB 8-14 to improve the dispersion in the aqueous liquor. Typical
emulsifiers being polyoxyethylated fatty alcohols, fatty acids, fatty acid tri-
glycerides, and glyceryl monofatty acids.
Another structure for the silicone glycol copolymer that has been de-
scribed [34] is
GMe 2Si(OSiMe 2)J0 (0SiMeG)40SiMe2G
where G = -(CH2 )J(OC 2H 4 ) 100Me. However, rather than formulating this
polymer with silica, it is formulated with a mixture of a polyoxyalkylene
214 Sawicki

glycol derivative such as C 8 H 17-(0C 2 H4 )JO(OC 3 H6 }J0Me and a nonionic


surfactant such as polyoxyethylene sorbitan monolaurate having 10 oxy-
ethylene units. The relative proportions being 50% by weight of the silicone
glycol copolymer, 30% by weight of the polyoxyalkylene derivative, and
20% by weight of polyoxyethylene sorbitan monolaurate. When the silicone
glycol copolymer was added to a high-temperature dye bath on its own it
only managed to control foam for 20 min with some staining on the polyester
fibers occurring, whereas when the same weight of the above composition
was used, foam was controlled for 53 min with no staining observed. As a
result such formulations are claimed to be particularly suitable for high-
temperature jet dyeing of polyester and other fibers. Usually 10-500 ppm
of the antifoaming composition is sufficient to control foam in a typical
dyebath. It should be noted that in the above silicone glycol copolymer struc-
ture the polyoxyalkylene groups are bonded to the silicon atom through a
Si-(CH 2 k-0-C linkage. This is because the copolymers are more hy-
drolytically stable than those in which the polyoxyalkylene groups are bonded
directly to the silicon atom through a Si-0-C linkage. The poorer hydro-
lytic stability is particularly noticeable in alkaline media where it manifests
itself as less durable antifoaming activity.
It is also possible to formulate the silicone glycol copolymers with the
reaction product of a silicone resin and a polyoxyalkylene [37]. A typical
formulation is 87% by weight of a silicone glycol copolymer with the structure
Me 3 SiO(Me2 SiO)w8 (MeGSi0) 1oSiMe 3
where G = -(CH2 ) 30(C 2 H4 0)dC 3 H 6 0)izOCCH3 , 13% by weight of a 50%
xylene solution of the reaction product of a resin, consisting of (CH 3 ) 3Si00 . 5
units and Si02 units in a ratio of 0.67: 1, and polypropylene glycol with a
molecular weight of 4100. The formulation was tested by adding 200 ppm
to a O. l % polyoxyethylene sorbitan monooleate solution at 86°C under high
shear. After running the recirculation pump for 6 min, the foam height was
5.1 cm when the silicone glycol copolymer was used on its own, whereas
with the reaction product present the foam only reached 2.5 cm. The time
for the foam to collapse at the end of the test after the recirculation pump
has been switched off was 12 and 5 s, respectively. Clearly the reaction
product enhances the antifoaming behavior of the silicone glycol copolymer.
More significantly though, when the reaction product was present, there was
no sign of the antifoam composition plating out and depositing on the surface
of the apparatus, whereas plating out did occur with the polymer on its own.
Plating out is the formation of an insoluble layer covering part of the surface
of the water that becomes visible when the recirculation pump has been
turned off. This is important when using these types of products as jet dyeing
foam control agents because plating out of the antifoam leads to staining
Antifoams for Textile Dyeing 215

and spotting of the fabric during dyeing. It has also been shown that it is
possible to incorporate these types of products into the dye carrier fornm-
lations to give a low-foaming textile dyeing auxiliary.
In order to achieve greater antifoam efficiency, these polymers have been
formulated with silicone antifoam compounds [32]. By formulating products
in this way it is possible to increase low-temperature antifoam efficiency
and achieve better foam control at the depressurization stage, while retaining
high-temperature performance and nonstaining properties. Such materials have
one major advantage over conventional silicone antifoam compounds de-
scribed earlier, and that is they are water dispersible. This means that the
textile dyer is able to simply disperse small quantities in cold water and add
them directly to the dye bath. The silicone glycol copolymers also ensure
complete dispersibility of the antifoam throughout the dyeing process. Fur-
ther, they retain their antifoaming ability even at the very end of the dyeing
cycle when the dye bath is cooled and depressurized. This stage is the most
critical with respect to foam formation, because foam is rapidly generated
by the sudden escape of dissolved air and the antifoam must not only limit
the amount of foam generated but also rapidly knock it down. These types
of formulated products are generally not quite as effective as conventional
silicone antifoam emulsions; however, their efficiency is sufficient to allow
the textile dyer to employ disperse dyes, in partially flooded jet dyeing ma-
chines, that otherwise would have to be excluded in view of their foaming
behavior. Furthermore, they can be used without the risk associated with
emulsion breakage. The use of such materials is now not just confined to
jet dyeing since progressively more applications are being developed that
need anti foam products that can withstand high-shear, high-temperature
conditions.
An example of such a formulated product is given below [38]. It consists
of four components:

A. Silicone glycol copolymer

where

B. Silicone resin copolymer which is the reaction product of a copolymer


consisting of Si0 2 units and (CH 3)JSi00 _5 and a hydroxylated poly-
oxypropylene polymer.
C. Silicone antifoam compound prepared from polydimethylsiloxane, pre-
cipitated silica, and siloxane resin.
216 Sawicki

TABLE 2 Antifoam Performance of Products Based on Silicone Glycol Copolymer

Test composition
Test number A B c D Foam profile
l 0 0 0 0 15 s to reach 7 .6 cm
2 100 0 0 0 5 min to reach 3.8 cm
3 50 50 0 0 5 min to reach 3.8 cm
4 94 5 I 0 3 min to reach 6.3 cm
5 90 5 5 0 5 min to reach 3.8 cm
6 90 5 3 2 30 s for the foam to reach 2.5 cm
and more than 5 min without the
foam reaching top of test vessel.
Source: Ref. 38.

D. Silicone glycol copolymer

Me 3Si (OSiMe 2) 10(0SiMeGh0SiMe3


where

The optimum proportions are 90 parts by weight of A, 5 parts of B, 3 parts


of C, and 2 parts of D.
The antifoam composition was tested in an 0.1 % aqueous solution of
polyoxyethylene sorbitan monooleate. The foaming solution was continu-
ously circulated at 86°C and the foam height and the time for the foam to
reach that height measured.
The bath was examined for signs of separation and plating out of the
components from the test composition. Ideally, there should be no separa-
tion or plating out whatsoever on the sides of the test vessel. This particular
test method exposes the antifoam composition to high amounts of agitation,
emulsification, and shear to simulate the conditions likely to be encountered
in a jet dyeing machine. The results are given in Table 2.
Though composition 6 did not have such good foam control performance
as compositions 2-5, the performance with respect to the intended end use
was rated excellent because the foam was controlled without any plating
being observed. The presence of the silicone glycol copolymer (ingredient
D) increases stability of the composition at the expense of antifoam perfor-
mance. This is yet another example where the foam control performance is
in part sacrificed by the formulator in order to minimize or eliminate a per-
formance defect.
Antifoams for Textile Dyeing 217

2. Nonsilicone-Based Antifoams
There are also foam control formulations that have been proposed that con-
tain neither polydimethylsiloxane fluid nor silicone glycol copolymers. They
are claimed to be effective under jet dyeing conditions and consist of com-
binations of ethylenebis (stearic amide), paraffin wax, hydrophobic silica,
nonsilicone hydrophobic oils, and a nonionic surfactant (39 ,40]. For ex-
ample [39], a dispersion of ethylenebis (stearic amide) in rapeseed oil is
added to a dispersion of polyethylene wax in mineral oil and stirred until
homogeneous. To this is added an emulsifier, which consists of an equi-
molar mixture of C 17H33CO(OC 2H4) 650COC 17H3 3 and C 17H33CO(OC 2H4k 50H,
hydrophobic silica, and finally a further addition of mineral oil. The resul-
tant product is a stable, fluid dispersion that can be diluted with water con-
taining 3.3% by weight of ethylenebis(stearic amide), 2.3% polyethylene
wax, 3.0% hydrophobic silica, 57.1% mineral oil, 26.1% rapeseed oil, and
8.2% emulsifier. To test the efficacy of this antifoam composition, pre-
washed polyester fabric was dyed in a Gaston County Mini-Jet machine with
a capacity of 650 L and 30 kg of fabric using the following dye liquor
ingredients:
0.95% C.I. Disperse Red (based on the weight of the substrate)
1 g/L Leveling agent, C 16_ 18 fatty alcohol condensed with 30 moles of eth-
ylene oxide
l g/L Dispersing agent, Turkey red oil
2 g/L Ammonium sulfate
200 mL Formic acid to pH 5
0.65 g/L Antifoam composition, as described above
450 L Demineralized water
17 kg Polyester, prewashed and prefixed
The goods-to-liquor ratio was 1:26 and the liquor circulation rate was 75
L/min. The liquor was heated from 60 to 126°C over a period of 2 hr. On
reaching 126°C dyeing was continued for 30 min, then the liquor was cooled
to 60°C over 30 mins. No foam formation was observed over the entire
process. Inspection of the dyed goods showed no spots or uneven areas due
to deposition of the antifoaming agent. Ethylenebis(stearic amide), paraffin
wax, hydrophobic silica, and hydrophobic oils are each known to have an-
tifoaming properties in their own right. However, this particular combination
gives unexpectedly good antifoaming properties under high-temperature
conditions. The compositions are claimed to be particularly suitable for use
in textile treatment liquors containing disperse dyes at high temperatures, in
the range 105-150°C, and under pressure.
An alternative foam control system which is also nonsilicone is a dis-
persion based on alkali earth metal salts of a fatty acid, dialkyl ester of an
218 Sawicki

unsaturated dicarboxylic acid, mineral oil, anionic and nonionic emulsifiers,


and ethylenebis(stearic amide) [41]. For example, 2.0% by weight of mag-
nesium stearate, 37% maleic acid esterified with 2-ethylhexanol, 35.5%
mineral oil, 24% nonionic and anionic surfactants, and 1.5% ethylene-
bis(stearic amide). Instead of the alkali earth metal salt of a fatty acid it is
also possible to use either the reaction product of a polyol, an anhydride of
an aliphatic dicarboxylic acid, and a polyalkylene glycol fatty acid adduct
or the reaction product of a polyol, an anhydride of an aliphatic dicarboxyhc
acid, and a polyalkylene glycol fatty alcohol adduct [42]. Such compositions
are claimed to be effective under the high-shear conditions encountered in
jet dyeing and so eliminate the necessity to continuously meter in the an-
tifoam to avoid spotting of fabric as in the case of antifoams based on sil-
icone emulsions.
As a result of the improvements made in formulation, it is now normal
to add these self-dispersible antifoaming agents to the dye tank along with
the other dyeing auxiliaries. For example, a typical dyebath composition for
dyeing polyester fabric in a jet dyeing machine would contain a disperse
dye, leveling agent, dye dispersant, ammonium sulfate, formic acid to pH
5, and antifoaming agent [39]. Ideally, little or no foam is observed over
the entire dyeing cycle and inspection of the dyed goods shows no spots or
uneven coloration.

IV. CONCLUSION
The textile industry in common with other manufacturing industries is striv-
ing to increase productivity and reduce energy consumption. At the same
time it is making every effort to meet increasing, environmental pressures.
One response to these pressures will be the development of wet dyeing pro-
cesses that involve the use of lower liquor ratios. This means that it will be
possible to dye with greater energy efficiency, because less energy is nec-
essary to heat the dye bath and to circulate the dye liquor when using lower
liquor ratios. Another advantage is that less effluent is produced that requires
processing. This implies that, in future, even higher concentrations of sur-
factants are likely to be present in dye baths, and consequently there will
be an even greater tendency for these baths to foam. Therefore, there will
be a need for even more effective, formulated, foam control agents that can
deliver antifoaming without affecting the quality of the dyeing process.
Foams arc also being used advantageously in textile processing, because
where a stable foam is created with an adequate distribution of textile treat-
ment systems within it, application of such systems to textile fabric can be
very cost effective as drying costs are dramatically reduced. For example,
in the foam finishing of fabric, products such as reactant resin formulations
Antifoams for Textile Dyeing 219

for cotton and polyester/cotton "wash and wear" finishes can now be ef-
fectively applied to fabric from an aqueous foam [43]. The low water content
of the fabric after foam application compared with conventional dip/nip pad-
ding operations gives higher productivity and lower energy usage during
drying and curing of the resin. The same principle has been used in the
dyeing of small articles such as garments and socks in rotating drum ma-
chines using dyestuffs dispersed in foams. This process, known as the San-
cowad process and developed by Sandoz, had some success a few years
ago. If such a process could be improved then new and different foam con-
trol agents would be required by the industry, because now foam needs to
be controlled after the dye has been applied rather than preventing it oc-
curring at all.

ACKNOWLEDGMENT
I thank Charlie Smith for his support in compiling the references at the outset.

REFERENCES
1. M. Daeuble, K. Oppenlaender, and R. Fikentscher (assigned to Badische An-
ilin-& Soda-Fabrik Aktiengesellschaft), US 3,830,627; August 20, 1974; filed
September 8, 1971.
2. R. H. Peters, Textile Chemistry, Vol. 3, Elsevier Scientific, Amsterdam, 1975.
3. G. Prazak, Am. Dyestuff Reporter 59: 44-46 (1970).
4. I. Slack, Canadian Textile J. 79: 46-51 (1979).
5. G. Weckler, Conference Proceedings, AATCC 1978 National Technical Con-
ference, pp. 140-145.
6. V. S. Salvin et al., Am. Dyestuff Reporter Nov., 23-24 (1959).
7. Encyclopedia of Chemical Technology, 3rd ed., Vol. 8, Wiley, New York,
1979, pp. 280-350.
8. W. T. Carpenter, Textile Chemist and Colourist 1(12):22-23 (1969).
9. A. Hodgson, Shirley Institute Publication S.35, 1979.
10. H. M. Tobin, Am. Dyestuff Reporter 68(9): 26-28 (1979).
11. I. R. Schmolka, J Am. Oil Chem. Soc. 59: 322-327 (1982).
12. R. E. Patterson, Textile Chem. Color 17(9): 181-184 (1985).
13. J. A. Colquhoun, Textile Industries (Atlanta) 137(8): 100-105 (1973).
14. Intl. Textile Bulletin, Dyeing/Printing/Finishing, 2nd Qtr., 30: 16 (1984).
15. H. Distler and D. Stoeckigt, Tenside Detergents 12(5): 263-265 (1975).
16. Intl. Dyer & Textile Printer 162: 351 (1979).
17. R.H. Via and S.C. Taylors (assigned to Milliken Research Corporation), US
4,428,751; January 31, 1984; filed March 29, 1982.
18. F. Redies, B. Redies, D. Tuerk, and C. Gille (assigned to Ciba-Geigy AG),
DE 2,943,754; May 14, 1980; filed October 30, 1979.
220 Sawicki

19. J. W. Keil (assigned to Dow Coming Corporation), US 3,984,347; October


5, 1976; filed December 19, 1974.
20. M. Daeuble, K. Oppenlaender, and R. Fikentscher (assigned to BASF AG),
DT 2,102,899; January 13, 1977, filed January 22, 1971.
21. A. Berger and H.U. Berendt (assigned to Ciba-Geigy Corporation), US
4,132,525; January 2, 1979; filed April 8, 1977.
22. Ciba-Geigy AG, GB 1,522,121; August 23, 1978; filed June 11, 1976.
23. H. Abel and A. Berger (assigned to Ciba-Geigy AG), DE 2,625,706; July 30,
1981; filed June 9, 1976.
24. H. Abel and A. Berger (assigned to Ciba-Geigy AG), DE 2,625,707; April
22, 1982; filed June 9, 1976.
25. H. F. Fink, H. Fritsch, G. Koerner, G. Rossmy, and G. Schmidt (assigned
to Th. Goldschmidt Aktiengesellschaft), GB 1,453,383; October 20, 1976;
filed September 5, 1974.
26. L. A. Rauner (assigned to Dow Coming Corporation) US 3,455,839; July 15,
1969; filed February 16, 1966.
27. R. D. Kulkarni, E. D. Goddard, and D. Kanner, Ind. Eng. Chem. Fundam.
16(4): 472-474 (1977).
28. R. D. Kulkarni and E. D. Goddard, Croatica Chem. Acta 50: 163-179 (1977).
29. R. D. Kulkarni, E. D. Goddard, and M. R. Rosen, J. Soc. Cosmet. Chem.
30(2): 105-125 (1979).
30. K. W. Farminer and P. B. Herter, Proceedings AATC Natl. Tech. Conf., 1974
Conf. Publication, Am. Assoc. Text. Color Chemists, pp. 337-342.
31. S. Ross, Rensselaer Polytechnic Institute Bulletin, Engineering and Science
series No. 63, Troy, New York, 1950.
32. D. N. Willing, Am. Dyestuff Reporter 69(6): 42-51 (1980).
33. A. N. Derbyshire and A. T. Leaver, J.Soc. Dyers Colour. 91: 253-258 (1975).
34. A. Abe (assigned to Shin-Etsu Chemical Co., Ltd.), US 4,042,528; August
16, 1977, filed March 25, 1975.
35. W. W. Cuthbertson (assigned to Imperial Chemical Industries Ltd.), US
3,700,400; October 24, 1972; filed May 3, 1971.
36. H. F. Fink, G. Koerner, G. Rossmy, and G. Schmidt (assigned to Th. Gold-
schmidt AG), DT 2,443,853; April 1, 1976; filed September 13, 1974.
37. J. A. Colquhoun (assigned to Dow Coming Corporation), US 3,912,652; Oc-
tober 14, 1975; filed December 17, 1973.
38. J. W. Churchfield (assigned to Dow Coming Corporation), US 3,746,653;
July 17, 1973;filedMay 15, 1972.
39. B. Danner (assigned to Sandoz Ltd.), GB 2,112,767; July 27, 1983; filed
November 22, 1982.
40. B. Danner (assigned to Sandoz Patent-GmbH), DE 3,505,742; September
12, 1985; filed February 20, 1985.
41. H. Abel, C. Guth, and H.U. Berendt (assigned to Ciba-Geigy AG), EP 263,069;
April 6, 1988; filed September 23, 1987.
42. H. Abel and R. Toepfl (assigned to Ciba-Geigy AG), EP 207,002; December
30, 1986; filed June 23, 1986.
43. J. J. Whiting and G.A. Richardson, Chemspec Europe 89 BACS Symposium
1989, pp. 11-13.
6
Foam Control in Detergent Products

HORST FERCH* AND WOLFGANG LEONHARDT Applied Technologies


Silicas and Silicates, Degussa AG, Hanau, Germany

I. Introduction 221
II. Foam Generation in Washing Machines 222
III. Surfactant Types Used in Laundry Detergent Formulations 226
IV. Types of Detergent Antifoams 227
A. Introduction 227
B. Soaps as antifoams 229
C. Nitrogenous antifoams 232
D. Phosphoric acid esters 235
E. Hydrophobed silica/ hydrophobic oil mixtures 236
V. Conclusion 263
References 263

I. INTRODUCTION
The development of foam with a product or during a process is sometimes
desired and sometimes not wanted at all. Shaving creams or the flotation of
ore are examples where the formation of foam is of some significance. Beer
which is not frothy is not very popular for aesthetic reasons.
On the other hand, foams can cause serious problems during many tech-
nical processes, for instance, in the purification of waste water, during cer-
tain applications in the coatings industry, or in the production of mineral

*Retired.

221
222 Ferch and Leonhardt

oil. It is therefore important to develop special additives or technical devices


which prevent excessive generation of foam.
Suppression or control of foam plays an important role during fabrics
washing processes. The development of the first detergents at the beginning
of the present century paralleled the development of washing machines. A
considerable amount of foam was generated by the detergent in those days
and was understood to be a sign of good cleansing power. It is still the case
nowadays under certain washing conditions.
Significant changes have taken place in the detergent sector over the past
few decades. The washing process has been continuously optimized in order
to satisfy the customers's wishes and to meet legislative demands, partic-
ularly during the last 20 years. The greatest changes have been made both
in the builder system and in the surfactant system. Here the builder is a
material which complexes divalent metal ions to prevent preciptation of in-
soluble salts of the anionic surfactant which is usually present in the sur-
factant system.
The tub-type machines formerly used in the European washing-machine
sector are gradually being replaced by drum-type washing machines. Due
to the nature of their design, the drum-type washing machines are very sus-
ceptible to the generation of foam with the result that much significance has
to be attached to this factor. Control of foam has therefore become an in-
tegral part of the work of those involved in developing new detergent
formulations.
The main aim of this article is to review developments in the formulation
and application of foam control compositions for heavy-duty laundry
detergents.
The practical part, however, does not intend to detail the performance
and composition of every suds-suppressing system for each powdered laun-
dry formulation. Emphasis is placed on their use in heavy-duty detergent
powders for European drum-type washing machines.
It is hoped that this contribution may open up profitable areas of research
and development for those who are not familiar with this special subject.

11. FOAM GENERATION IN WASHING MACHINE


A closer look at the factors determining the washing process will allow a
better understanding of the problems related to the development of foam.
The "laundry," or articles to be washed, the method of washing, and the
detergent are all integral parts of the washing process. These partners are
well coordinated in order to ensure that optimum results are attained during
the cleansing of the laundry.
Foam Control in Detergent Products 223

I II III

FIG. 1 Schematic drawing of different automatic washing machines


I = agitator-type (U.S.A); II = pulsator-type (Japan); III = drum-type (Europe)
a = timer; b = basket; c = agitator; d = outer tub; e = motor; f = pulsator;
g = thermostat; h = detergent dispenser; i = drum; k = carrier baffle;
I = heating coil. (From Ref. l.)

Washing habits vary considerably around the world, and this is mani-
fested in the widely differing methods of washing. In underdeveloped coun-
tries, for example, washing of laundry is usually done by hand. Here copious
amounts of foam are associated by consumers with successful cleaning per-
formance by the detergent. Detergent formulations for such applications should
therefore be high foaming.
Washing machines are used almost without exception to clean the laundry
in the industrial nations of Europe, North America, and in Japan. Typical
designs of such washing machines are shown in Fig. l [ l].
There are two basic designs of washing machine: the tub-type washer
used in North America and Japan and the European drum-type washing ma-
chine [2]. Wash conditions used with these different types of machine in
the three markets are summarized in Table 1.
North American and Japanese washing machines generally tend to be top-
loaders, because they are invariably filled from above. The outer tub is made
out of enamel-coated tin or plastic (Japan) and contains the perforated wash-
tub which turns around a vertical axle. In the American washing machines
the mechanical action is conveyed by means of a rotating agitator which is
situated vertically in the middle of the washtub. In contrast, Japanese ma-
chines are equipped with a flat rotating disklike impeller fitted to the base
224 Ferch and Leonhardt

TABLE 1 Washing Habits and Conditions in Different Parts of the World

Washing conditions USA/Canada Japan Western Europe


Washing machine Agitator type Impeller type Drum type
Heating coils No No Yes
Amount of wash Extra small: approx. Low: 30 Low:J6-20
liquor (1) 35
Medium: approx. 50 High: 45 High: 25-30
Large: approx. 65
Extra large: approx.
80
Total water con-
sumption (I)
(regular heavy cycle) 140 150 120
Wash liquor ratio 1:15-1:30 1:20-1:30 1:5-1:25
Washing time (min) 10-15 5-15 60-70 (90°C)
20-30 (30°C)
Washing, rinsing and 20-25 15-35 100-120
(90°C)
Spinning time (min) 40-50 (30°C)
Washing temperature Hot: 50 10-40 90
(oC)
Warm: 27-43 60
Cold: 10-27 40
30
Wash cycle/month 22 60 18
Water hardness Relatively low Very low Relatively high
(ppm CaC03) 100 50 250
Method of detergent Mostly by hand into Mostly by Dispenserb
addition the basket hand into the
basket
detergent dosage• 1.5; 1.3; 7-10;
(g/L)
(g/kg fabric) 35-50 30-40 60-80
'In USA and Japan without bleaching components
"Pre-wash and main-wash cycle and fabric softener (and chlorine bleach in Southern Europe)
Source: Ref. 2.

of the washtub; this ensures that the wash liquor is kept in motion and leads
to a through soaking of the laundry.
Neither of these two types of washing machine is equipped with any kind
of heating element. In Japan, the washing is done at the temperature of the
drinking-water supply as it comes from the tap. American washing machines
are directly connected to the hot-water supply in the house and a cold-water
Foam Control In Detergent Products 225

connection allows the temperature of the water to be regulated according to


the wash program chosen. Detergent is measured by hand directly into the
washtub. Here, the following order is recommended: detergent, laundry, and
then the water.
Foam-related problems are not to be expected with these Japanese and
American tub-type washing machines because of limited mechanical agita-
tion leading to low rates of air entrainment, low washing temperatures, and
low detergent concentrations (see Table 1) [3].
Two different designs of European drum-type washing machines have
established themselves. These are top-loaders and front-loaders. The former
are filled from above through a hatch in the drum, and the latter are filled
directly into the open drum through a door at the front of the washing machine.
The washing drum is perforated and revolves around a horizontal axle in
varying cycles, reversing at intervals; it is situated lengthwise within a re-
ceptacle made out of stainless steel. This longitudinal outer tub is just slightly
larger than the actual washing drum and is only about one third full of wash-
ing liquor during the washing process. Inside the drum are carrier baffles
which lift the laundry lengthwise as the drum revolves, causing the laundry
to be dropped back down into the water from the highest point. This results
in an intensive mechanical agitation and ensures the close contact of the
laundry with the washing liquor.
Depending on the wash program chosen the wash liquor, which is cold
at the outset, is heated to and maintained at the desired temperature by means
of built-in heating coils. The maximum temperature lies at approximately
95°C. The machine is programmed to automatically flush detergent out of
the dispenser into the washing machine. This detergent dispenser contains
three to four compartments into which the measured dose of the detergent
for the prewash and main-wash cycles is filled together with any aftertreat-
ment aids, such as fabric softeners and chlorine bleach. The latter is only
used in some parts of southern Europe.
The design of European drum-type washing machines conspires with high
washing temperatures and high detergent concentrations to produce rela-
tively severe foam control problems. An important design factor is the nature
of the mechanical agitation. This is responsible for entrainment of copious
amounts of air and the development of undesirable foam. Air entrainment
occurs in a number of different ways:
When the laundry is lifted out of the washing liquor by the carrier baffles
situated inside the drum
When the laundry falls back into the washing liquor
When the washing liquor flows through the perforations in the washing drum
The amount of foam generated must be regulated in order to attain good
washing results. Too much foam results in insufficient contact between the
226 Ferch and Leonhardt

TABLE 2 Formulations of Detergents (Heavy-Duty Powders)

USA/Canada/ Western
Ingredient Australia Japan Europe

Surfactants 8-22 19-25 8-13


Builders 20-35 18-30 20-45
Bleaching agents a a 15-35
Auxiliary substances
Foam boosters 0-2
Foam depressants 0.3-5
Bleach activators 0-3
Bleach stabilizers 0.2-0.5
Antiredeposition agents 0-0.9 0-2 0.5-1.5
Anticorrosion agents 5-15 5-15 2-8
Enzymes b 0-0.8 0-0.5
Optical brighteners 0.1-0.8 0.15-0.8 0.1-0.8
Fabric softeners 0-5 0-5
Perfumes c c c
Fillers and water Balance Balance Balance
'Bleaching is carried out separately with a liquid chlorine bleach during the rinsing cycle.
bOnly very few detergents in the US contain enzymes up to 0.75%.
'Perfumes are present.
Source: Ref. 4.

laundry and the washing liquor. The entire mechanical washing process is
impeded because the foam prevents the laundry from being able to fall freely
into the washing liquor. In extreme cases foam and washing liquor may even
flow out of the top of the machine.
The need to regulate the amount of foam generated during the washing
process is, therefore, solely a problem incurred with European drum-type
washing machines and is absolutely essential if optimum washing results are
to be achieved.

Ill. SURFACTANT TYPES USED IN LAUNDRY


DETERGENT FORMULATIONS
The washing conditions described in Sec. II have resulted from decades of
development and reflect differences in textiles, wearing habits, washing ma-
chines, water hardness, washing temperatures, and duration of wash cycles.
The "chemical" part of the washing process, i.e., the detergent, takes all
these factors into consideration. It leads to different detergent compositions
as shown in Table 2 [4].
Foam Control in Detergent Products 227

Surfactants are the most important ingredients in the detergent and have
gained further significance over the past 40 years. The choice of surfactants
for detergent applications depends on a series of factors; the main ones are
Good detergency (cleansing) properties
Good handling
The manufacturing process for detergents
Economy
Toxicological safety
Ecological safety
Two main classes of surfactants are used in detergent applications: anionic
and nonionic surfactants. The former group of surfactants is used most due
to a favorable cost/benefit ratio. Good solubility, good soil removal and
good dispersion effects render these anionic surfactants a valuable compo-
nent of modern detergents [5,6]. Following a reduction in washing temper-
atures, the development of new textiles and the replacement of phosphates
in detergents, nonionic surfactants have gained considerable importance in
the past 10-15 years. Even at lower concentrations they are comparable to
anionic surfactants with respect to cleansing power, especially the dispersion
of soil.
Surfactant mixtures consisting of anionic and nonionic surfactants are
generally employed in detergents. Amphoteric surfactants are also used for
special applications and the special properties of cationic surfactants, which
are often applied in combination with layer silicates, produce the fabric soft-
ening effect.
Some of these mixtures consist of several surfactants and tend to display
foaming behavior which is not satisfactory for practical purposes. Despite
the fact that the structural parameters of the surfactants are known, it is
an extremely complex matter to relate these to the foaming properties of
a surfactant. In general, this cannot therefore be done, and the foam be-
havior of surfactants in a detergent cleaning context in usually assessed
empirically.

IV. TYPES OF DETERGENT ANTIFOAMS


A. Introduction
Early development of antifoams for laundry detergents was hampered by
ignorance of the physicochemical processes involved. Consequently an em-
pirical approach had to be adopted. Even today, it is difficult to establish
clear structure-effect relationships concerning the foam-inhibiting properties
of a given antifoam system. The desirable properties of an antifoam can be
228 Ferch and Leonhardt

TABLE 3 Classification of Antifoam Substances

Type Substance

Carboxylates Fatty acids


Fatty acid esters
Soaps
Phosphates Monoalkylphosphoric acid esters
Dialkylphosphoric acid esters
Fluoroalkylphosphoric acid esters
Nitrogene-containing Melaminresins
compounds Amides
Amines
Ureas
Hydrocarbons Mineral oils, -waxes
Paraffins
Synthetic oils, -waxes
Edible oils, -waxes
Natural waxes
Microcrystalline waxes
Organic silicon compounds Polydialkylsiloxanes
Silicone resins
Fluoro compounds Partially /totally fluorinated alcohols and carbonic acids

listed as follows:
High efficiency at a low concentration
Applicability within a wide temperature range and with different washing
times
Insensitivity to the water hardness
Good price/performance ratio
High stability in the detergent
Toxicological and ecological safety
A wide range of compounds has been tested, some of which have proved
to be suitable for practical use. These compounds are classified according
to chemical type in Table 3. This list is not claimed to be complete.
The composition of antifoams has become more complex in the past 15
years. In part this has been due to the use of particulate substances, such
as hydrophobed silica in admixture with hydrocarbons or polydimethylsi-
loxanes. In many such systems the addition of hydrophobed silica improves
the antifoam efficiency markedly (see Chapter 1).
Today's antifoams can, therefore, be classified as silica bearing and silica
free. This fact will be considered in the following overview, and the most
Foam Control in Detergent Products 229

"Overfoaming•

5 Foam in the
dispenser tube

4 4/4 of the front glass

3 3/4 of the front glass

2 2/4 of the front glass

114 of the front glass

o No foam

FIG. 2 Description of the foam rate system introduced by Schmadel [ 11].

important developments discussed on the basis of the patent literature re-


ferred to. Practical examples (foam curves) will be used to illustrate certain
correlations. The measurement is carried out in a front-loading washing ma-
chine (W433, Miele/Germany) which has an inbuilt electronic mechanism
to measure via conductivity the amount of foam generated during the wash-
ing process. The foam curves obtained are evaluated on the basis of the
foam-grading system proposed by Schmadel [7], which is illustrated in Fig.
2.

B. Soaps as Antifoams
Organic carboxyl compounds, such as carboxylic acids and their alkali salts
(soaps) were the first compounds to be used as antifoams in low-foaming
detergent formulations. Early work on the antifoam behavior of such ma-
terials was due to Harris [8] and Peper [9]. The effect of soap on the foam
of sodium dodecylbenzenesulfonate and cetyl sulfate was examined as a
function of the pH and water hardness (calcium ion concentration). The ex-
periments showed that soap or fatty acids have an antifoaming effect only
when the conditions favor the formation of slightly soluble calcium soaps
and when there is no formation of mixed surface films consisting of calcium
soaps and other surfactants. Islands of insoluble and solid calcium soaps are
supposed to form, which, due to their inflexible nature, destabilize foam
films [10].
Both water hardness and builder systems have a decisive influence on the
effectiveness of soaps. As can be seen from Fig. 3, there must be an ade-
quate calcium ion concentration available to enable sufficient calcium soaps
to form [11]. This is also the cause of problems with excessive foam for-
mation in so-called soft-water areas when soap is used as an antifoam. More-
230 Ferch and Leonhardt

Foam rate

2
* deionized water
0 0.54 mmol ca2 •11
9 2.9 mmol ca 2 •11

20 30 40 50 60 70 80 90 100
Temperature [°C]

FIG. 3 Antifoam behaviour of sodium behenate as a function of water hardness.


A: deionized water; B: 0.54 mmol Ca2 +/L; C: 2.9 mmol Ca2 + /L. Detergent: heavy-
duty detergent with 3. I% linear alkyl benzene sulfonate (LAS); dosage: I 00 g pre-
wash and 100 g mainwash; washing machine: Miele W 433 (drum-type, front-loader);
wash load: 3.5 kg clean cloth. (From Ref. 15.)

over, if soap is the only antifoam, then only builder substances can be used
which have relatively low stability constants. In the presence of the builder,
the calcium ion concentration must be high enough to allow formation of
the calcium soap. This is the case with sodium tripolyphosphate. However
problems may be encountered with, for example, nitrilotriacetic acid (NT A)
or new builder combinations in today's phosphate-free detergents. An ex-
ample showing the effect of NTA is given in Fig. 4 [ 12].
Soaps with chain lengths of 12 to 22 C atoms [ 13, 14] are usually em-
ployed. Long-chain soaps (C 20 22 ) have however been found to be particu-
larly effective for wash temperature above 60°C. Low foam profiles are pro-
duced with these soaps when sodium tripolyphosphate is used as builder. A
typical soap blend and/or the fatty acid blend from which it is derived, has
the following composition: 5-15% myristic acid (C 14 ), 25-35% palmitic
acid (C 16), 15-25% stearic acid (C 18 ), 20-30% arachidic acid (C 20 ) and 10-
20% behenic acid (C 22 ).
Precipitation of finely divided calcium soap is facilitated if the sodium
soap dissolves readily at the relevant temperature. That is the wash tem-
perature should exceed the Krafft temperature of the sodium soap. Therefore
Foam Control in Detergent Products 231

Foam rate

20 30 40 50 60 70 80 90 100
Temperature [°C]

FIG. 4 Foaming behavior of soap-regulated detergents based on 40% NT A (0) or


40% tripolyphosphate (*); dosage 7 .5 g/L. (From Ref. 12.)

at low wash temperatures only short-chain-length soaps are effective. How-


ever, the solubility of the corresponding calcium soap precipitate in the pres-
ence of STP increases with increasing temperature rendering it ineffective.
Foam control at higher temperatures then requires the presence of long-chain-
length soaps the calcium soaps of which are still insoluble.
Figure 5 shows a comparison between tallow soap and behenic soap [15).
There is a distinct reduction in the efficiency of the tallow soap just before
the boil wash temperature is attained, and the machine overfoams. The foam
remains at a low level when behenic soap is used. Foam formation at the
beginning of the washing process is attributable to the low solubility of both
sodium soaps so that the rate of formation of finely divided calcium soap
antifoam is inhibited.
The concentration of soap in detergent ranges from 3% to 6%. Processing
is effected together with the other detergent constituents in the detergent
slurry which is homogenized in a crutcher and dried in a spray tower. A
loss in activity when the detergent is stored, such as occurs in the case of
the antifoam systems still to be discussed, is not observed.
The main disadvantages for soap are the relative weekness of the anti foam
ability (as revealed by the high levels required) and its dependency on water
hardness. There is also the danger that calcium soaps are deposited on the
232 Ferch and Leonhardt

Foam Rate Temperature {°C)


5-,-------------------------~100

0 -r-----r----.,-----..------,-----r-----.----'- 0
0 10 20 30 40 50 60
Time (min)

FIG. 5 Comparison between Tallow (C, 6_, 8) and Behenic (C 20_ 22 ) soap in a STPP-
reduced detergent (8% LAS, 4% nonionic surfactants, 16% zeolite and 22% STPP)
with a soap content of 3.5%. Washing machine: Miele W433 (drum-type, front-
loader); dosage: 8.5 g detergent/L; water hardness: 2.85 mmoi Ca2 + /L.

fabric contributing to organic incrustations and a high degree of yellowing.


Some improvements have been claimed to have been achieved by selec-
tive modification of the fatty acid molecules from which the soaps are de-
rived. These include fatty acids with side chains [16], isomeric fatty acids
[17] and di- or trimerized oleic acids [18].
The amount of soap required for foam control can be reduced if nonionic
surfactants, characterized by a high degree of ethoxylation, are used as co-
surfactant in the laundry detergent [19,20]. As a result of this a larger quan-
tity of anionic surfactants can be used to yield a noticeable improvement in
the cleaning efficiency.
C. Nitrogenous Antifoams
One class of effective antifoams comprises nitrogenous compounds such as
amines, urea derivatives, amides, and trialkylmelamines. Only the two latter
compounds have gained a certain significance and should, therefore, be dis-
cussed. Both mono and diamides, (l) and (£), are used as antifoams.
Monoamides (21,22]
R-CONHR 1 (l)
Foam Control In Detergent Products 233

Diamides [23-26]

R 1-C-N-R-CH 2-N-C-R3
I I I I
0 R2 R4 0
R 1-CH 2- N-C-C-N-CH2-R3
I I I I
R 2 0 0 R4
R 1-CH 2-N-C-R-C-N-CH2-R3
I I II I
R2 0 0 R4
R = c,_9; R,,3 = C11-23; R2,4 = H, cl-24
It has been claimed that monoamides (l) are highly efficient antifoams. In
a typical, highly foaming detergent (17% alkylbenzene sulfonate, 3% of a
nonionic surfactant, 45% tripolyphosphate, 8% sodium silicate, 1% CMC,
22% sodium sulfate, and 4% water) good results can apparently be achieved
with as little as 1% octadecanamide. These materials may be introduced
through the slurry. Thus after neutralization of the alkylbenzene sulfonic
acid with caustic soda a well-homogenized premixture of nonionic surfactant
and the amide are added and homogenized. Other powder constituents are
then added. The slurry prepared in this way is then spray-dried.
Antifoam diamides G) are substances which can have high melting points
(mp), for example, distearyldiamide: mp 123-137°C. They are exclusively
used in combination with additives such as polysiloxanes, oils, waxes, other
hydrocarbons, or even soaps. They are only effective when they are finely
dispersed in the detergent solution so that they are quickly transported to
air-water surfaces of the foam. The desired foam profile depends on the
chain length [23]. Thus if R 1 and R 3 have chain lengths of C 12_14 , then the
resulting compound has an optimum antifoam range at 20-60°C. Longer
chains (C 18_ 22) have an optimum range between 60 and 100°C. With dosage
levels of the antifoam compound in the range of 0. l % to 3% , this antifoam
system is highly efficient.
The advantage of diamides is that their effect is not dependent on the type
of surfactant or the water hardness. Moreover, relatively low feed concen-
trations require different processing methods. Processing in a detergent slurry
together with other constituents is, in fact, possible, but leads to activity
losses, since the required disperse distribution is not achieved. A better and
more effective method is the production of a homogeneous premixture by
means of dispersion processes with a high shearing action, and subsequent
further processing of this dispersion by spraying onto the washing powder
[23-25,27] or by deposition on a carrier substance [25,27]. An example of
234 Ferch and Leonhardt

an effective antifoam system consists of diamide and a polysiloxane which,


after deposition on a sucrose or gelatinized starch carrier, is protected by
coating with paraffin wax and thus has a long-term efficiency [24].
Combinations of 5-20% distearyldiamide with 20-50% soft/hard par-
affin (mp 20-55°C) and 25-60% microparaffin (mp 60-80°C) have been
found to be very effective, 0.1-0.3% of this combination being sufficient
for effective antifoam action in a detergent formulation [27). Approximately
10% of this antifoam are deposited 127) on a carrier (light sulfate). Subse-
quent coating is not necessary due to setting of the wax matrix. Particular
attention is drawn to the good long-term storage stability in the washing
powder.
Another group of nitrogenous antifoams Q) is formed by the trialkyl-
melamines [28-36):

R-HN......._ ......N~ ....-NH-R


c c
II I
N-......C':;:::-N
I
NH-R R = C7_22

Melamines of cyanuric chloride and fatty amines are available [37-391. These
are characterized by their insolubility in water and, depending on the sub-
stitution, by melting points of between 30°C and 90°C. As with the di-
amides, these arc not processed through slurry. The wash active substances
combined with the high temperatures would emulsify the melamine too much
and cause it to become inactive.
A two-stage process has been found to be practicable. First, a liquid to
slightly pasty emulsion is produced from melamine resin, emulsifier, and
other additives. This emulsion is sprayed directly onto the washing powder
in a second stage, or onto a carrier, which is then incorporated into the
washing powder [33,34]. The advantage of the latter variant is that the car-
rier "dilutes" the active antifoam, as a result of which the anti foam is dis-
tributed uniformly and more homogeneously in the washing powder. At the
same time, this measure improves storage stability.
Examples of the abovementioned emulsions are listed as follows [331:

Example 1 Example 2
18% melamine 40% melamine
3% C 16_ 18 fatty alcohol with 10 mol EO 20% fatty alcohol (see left)
9% C 10 fatty acid (capric acid) 40% paraffin oil
70% water
Foam Control in Detergent Products 235

Foam rate

* C0/C12 - melamln
X C16/C10 - melamln
\7 1 : 1 - mixture

20 30 40 50 60 70 80 90 100
Temperature (°C]

FIG. 6 Antifoam behavior of melamines with short-chain and long-chain alkyl groups
in comparison to their l: I mixture. Detergent: heavy-duty STPP-based detergent;
detergent dosage: 100 g main wash; melamin dosage: 2.6%; wash load: 3.5 kg clean
cloth; water hardness: 2.8 mmol Ca 2 + /L.

When the residual alkyls are properly selected, a desired, temperature-in-


dependent foam profile can be obtained (Fig. 6). Melamines with short-
chain residual alkyls are effective up to about 70°C, while melamines with
long-chain alkyls only begin to have an antifoam effect above about 50°C.

D. Phosphoric Acid Esters


A class of compounds used for antifoam purposes is formed by phosphoric
acid esters(~:). These are comparable to the fatty acid/soaps [40] in forming
slightly soluble or insoluble salts with multivalent cations. The mode of ac-
tion of the phosphoric esters is therefore probably similar to that of fatty
acids in that insoluble precipitates of calcium salts are usually involved.
0
I
HO-P-OR1
I
OR 2
R, = C9_20;
236 Ferch and Leonhardt

The disadvantage of higher manufacturing costs relative to the soaps is com-


pensated to a far-reaching degree by greater effectiveness, Thus phosphoric
acid esters are effective in dosages of as little of 0.5-3% [41,42].
Efficiency can be further increased by combining phosphoric acid esters
with other additives or by modifying and optimizing the method of pro-
cessing. Above all, the disadvantage already encountered with the fatty acids
of efficiency being dependent on the water hardness, should be taken into
consideration. Nonionic surfactants can be used as additives [43-45]. Thus
for example detergent mixtures which contain 0.1-0.5% C 16_ 24 phosphoric
acid esters and 1-3.5% tallow alcohol with 11 mol ethyleneoxide are de-
scribed [46]. The phosphoric esters and the nonionic surfactants are melted
together, homogenized and then sprayed onto the washing powder. Com-
binations with soap [47] and hydrocarbons, such as mineral oils, or waxes
of synthetic, mineral, vegetable or animal origin [48-52), are also described.
0
II
HO-P-(EO)n-OR 1
I
x (2)
X = OH, R 20(EO),,; m,n = 0,1-6

The use of modified phosphoric esters (2) which contain a polyoxyethylene


chain between the residual alkyls and the phosphoric acid ester group has
been described [53,54]. The direct employment of the slightly soluble cal-
cium salts of these alkyl phosphoric esters can be particularly advantageous.
These salts may be dispersed in hydrocarbon wax. To obtain optimum ef-
ficiency the particle size of the calcium salts should be between 0.1 and 0.5
µm [49]. The melting point of the waxes employed has a considerable in-
fluence on the antifoam efficiency and should be below the maximum wash-
ing temperature. The antifoam is either hot-sprayed, in the form of the dis-
persion described above, onto the finished washing powder or processed
together with other powder components to form antifoam agglomerates. This
can be done using known processes, such as extrusion or pan agglomeration.
Use of phosphates in detergents is increasingly controversial, especially
in Europe. Less and less interest is therefore being shown in phosphoric acid
esters, and these are disappearing from detergent formulations.

E. Hydrophobed Silica/Hydrophobic Oil Mixtures


Silicone- and mineral oil-based antifoams activated by the addition of silicas
represent one of the most important class of antifoams employed today. Here
neither the silica nor the oil is effective alone. The mixtures, however, ex-
hibit synergy provided the silica is rendered hydrophobic. This type of be-
Foam Control in Detergent Products 237

Sand

~N_a~2c_o~3~1N_a_o_H~J L'~~~~~c~~~~~
..!. l.

1
Waterglass (Ferro)silicon

HCl/Cl 2

1
H SO
2 4
C/O 2 Si Cl
4

Hz'02

l l
Silicagel Precipitated silica Arc silica Fumed silica
(Hydrogel,
Cl 2-silanes Cl 2-silanes
Aerogel,
Silicone oils
Xerogel) Silazanes
Waxes

surface-treated surface-treated
Precipitated silicas Fumed silicas
FIG. 7 Manufacturing of different silicas from quartz sand.

havior is also shown when, for example, intrinsically hydrophobic particles


such as calcium alkyl phosphate or ethylenebis (stearamide) are mixed with
mineral oils. The mechanistic aspects of this behavior are reviewed in detail
in Chapter 1 .
1. Silica Manufacture, Properties, and Application
The synthetic silicas frequently used today are tailormade )'-ray amorphous
products. Depending on the manufacturing process, these can be divided
into the following groups [55-57]:
Pyrogenically or thermally produced silicas (fumed silicas)
Wet-chemically produced silicas (precipitated silicas)
Aftertreated silicas
Silica sols occupy a special position, since they are not present in a partic-
ulate state. They can, however, be considered as the precursor to silica gels.
A raw material common to all synthetically manufactured silicas is quartz
sand of as high purity as possible. For manufacturing wet-chemically pro-
duced silicas, water glass is prepared, either thermally or hydrothermally,
from quartz sand. By contrast the basis for the pyrogenic products is silicon
tetrachloride, produced from silicon or ferrosilicon (Fig. 7).
Only the fumed and precipitated silicas have acquired significance for use
as antifoams. Fumed silicas are highly disperse silicas which form at high
238 Ferch and Leonhardt

Hydrogen HCl-absorption

Oxygen (dry air)

SI-tetrachloride b

d fumed
Silica

FIG. 8 Scheme of the production process of Aerosil (fumed silicas): a = evapo-


rator; b = mixing chamber; c = burner; d = cooler; e = cyclone; f = dcacidizing;
g = silo.

temperatures due to coagulation from the gas phase. This process is de-
scribed as flame hydrolysis (Aerosil process [58-60]). A homogeneous mix-
ture of silicon tetrachloride, steam, hydrogen, and dry air is burnt. The en-
ergy and water necessary for hydrolysis is supplied by the oxyhydrogen flame.
The silica is separated in cyclones and neutralized with water vapor and air
in a fluidized-bed reactor. The product which has a bulk density of 20 g/L
is pneumatically conveyed to packaging machines, where it is simulta-
neously compressed to 60-100 g/L (Fig. 8).
Precipitated silicas are formed by the reaction of alkali silicate (water
glass) with a mineral acid (1).

(I)

By varying the chemical and physical conditions during prec1p1tation, as


well as the method of filtration, drying, and grinding, a number of different
products can be manufactured, whose characteristics fulfil the requirements
for certain fields of application. The type of precipitation process, in which
further silica is deposited on the primary particle formed, thus leading to a
growth in the particle size, is essentially responsible for the fact that only
discontinuous manufacturing processes have achieved industrial-scale
significance.
Foam Control in Detergent Products 239

Waterglass

Sulfurlc acid Water Steam

....
Wastewater

Precipitated
Silica

FIG. 9 Scheme of the production process of precipitated silicas: a = precipitation


reactor; b = filtering apparatus; c = drying; d = intermediate product silo; e =
grinder; f = silo; g = densifier.

A water-glass solution is mixed with acid, usually sulfuric acid, under


defined temperature and stirring conditions in large mixing vats (Fig. 9)
[61,62]. It is also possible to use carbonic acid and hydrochloric acid [63,64).
The suspension which results is filtered off (usingly chamber presses, plate
and frame presses, automatic filter presses, or band and rotary filters) and
washed free of salt. The filter cake (solids content 15-25%) is dehydrated
on band, turbine, revolving, or rotary driers, or also by spray-drying after
prior liquefaction of the filter cake [65,66).
The properties of the amorphous silicas manufactured according to the
processes described above, are determined by the particle size, particle
structure, and the chemistry of their surface, which is characterized by the
presence of silanol groups Si-OH. These groups are responsible for the
hydrophilic nature of silica surfaces. These surface silanol groups can be
chemically reacted with various reagents to render the silica surface
hydrophobic.
In the case of fumed silicas this is usually effected with chlorosilanes in
a fluidized bed reactor at 400°C [67). Precipitated silicas can be hydro-
phobed with alkylchlorosilanes in the precipitated suspension. In this case
filtration, washing and drying are followed by tempering to 300-400°C
[68,69). The hydrophobing mechanism is illustrated in Fig. 10. In addition
to chlorosilanes, silazanes, and, particularly in the case of precipitated sil-
240 Ferch and Leonhardt

CH 3 CH 3
0
"\ I \ I
H H CH 3 Cl CH3-Si Si-CH3
I I ~ / I \
0 0 SI 0 0
/
I I CH 3 "'Cl I I
Si

FIG. 10 Reaction between dimethylchlorosilane and the silanol groups of silica


surfaces.

icas, polydimethylsiloxanes (silicone oils) are used [70]. Treatment with sil-
icone oils, which is also known as "in situ hydrophobing" has achieved
particular significance with respect to the use of silicas in antifoams, since
the silicone oil is not only a hydrophobing agent but also an important com-
ponent of the antifoam formulation.
A less important method of surface modification is the so-called coating
process. Here the hydrophobic agent is only physically adsorbed on a silica
surface so that a chemical reaction with the silanol groups is not involved.
Fused-on waxes are used [71] for this purpose, or, in the case of precipitated
silicas, wax dispersions [72].
The essential characteristics of the fumed and precipitated silicas are shown
in Table 4. The most pronounced differences are the aggregate and agglom-
erate size and the lack of porosity of the fumed silicas, which only have an
external surface. With aftertreated silicas, a further characterizing parameter
is the degree of hydrophobicity.
In recent years, hydrophobic, precipitated silicas have proved to be ex-
tremely effective in silicone- and mineral oil-based antifoams, special an-
tifoam silicas having been developed [56]. The essential characteristics are
the degree of hydrophobicity, agglomerate size and specific surface (Table
5). A favorable influence is possibly exerted on the antifoam efficiency by
the surface roughness. However, the degree of hydrophobicity has the most
decisive influence on the efficiency. This can be determined either by mea-
suring the contact angle at the air/water surface or by measuring the "meth-
anol wettability."
Foam Control in Detergent Products 241

TABLE 4 Comparison of Important Properties of Synthetic Amorphous Silicas

Silica type
Parameter Fumed Precipitated

BET-surface [m2/g] 50-600 30-800


Primary particle size [nm] 5-50 5-100
Agglomerate size [µm] a 1-40
Density [g/cm3] 2.2 1.9-2.1
Tapped density [g/l] 50-100 50-500
(DIN 53 194)
Drying loss [%] <2.5 3-7
(DIN 53198)
Ignition loss [%] 1-3 3-7
(DIN 55921)
pH value 3.6-4.3 5-9
(DIN 53200)
DBP adsorption 250-350 150-320
(cm3 /100 g)
Percentage inner surface Nil Low
Structure of aggregates/ Chainlike Slightly aggregated,
agglomerates agglomerates spherical particles
Thickening effect Very strong Existing
Average pore size (nm) Nonporous >30
'Fumed silicas do not have a defined agglomerate size

The contact angle is measured on pressed silica tablets (Fig. 11). A drop
of water is placed on the surface of the pressed silica and the contact angle
determined optically by means of a goniometer. Hydrophobic surfaces are
characterized by an angle of >90°.
Methanol wettability is a measurement which relies upon the reduction
of the surface tension of water by addition of methanol. If the proportion
of methanol is high enough, the critical wetting tension of a hydrophobed

TABLE 5 Properties of Antifoam Silicas

Parameter Requirement

Hydrophobicity Contact angle >90°


Methanol wettability >50%
Particle size 0.2-5 µm
Specific surface BET >50 m 2 /g
pH >8
242 Ferch and Leonhardt

0 0
Angle< 90 Angle> 90

Oil Oil

FIG. 11 Definition of the contact angle of water droplets placed on the surface of
hydrophilic or hydrophobic silicas (pressed tablets immersed in oil).

silica can be achieved in that the material is wetted by the solvent. Methanol
wettability therefore simply involves measurement of the proportion of hy-
drophobed silica dispersed into a water/methanol mixture as a function of
the composition of the mixture. Typical plots are shown in Fig. 12.
2. Hydrophobed Silica/Hydrocarbon Antifoams
The range of hydrocarbons employed in antifoams includes solid and liquid
products of mineral (petrochemical), natural or synthetic origin. The term
hydrocarbon is not quite precise in this context and only serves to simplify
and facilitate the following explanations.
Hydrocarbon waxes are characterized by melting points above 20°C and
low saponification values of ~60. Use in detergents requires that the ma-
terial melt and disperse in the wash. This means that the melting point must
be below wash temperatures which are always :S95°C and often <50°C. The
optimum melting point range is therefore 30-70°C. An important criterion
is the type of "crystallinity." Microcrystalline waxes as described by Warth
[73) are very effective. These waxes have smaller crystals than paraffin waxes,
which is attributable to the higher portion of branched-chain hydrocarbons
and to a higher molecular weight.
Vaseline (petroleum jelly) with a melting range of 35-50°C has found
wide application. Synthetic waxes of the Fischer-Tropsch type (named after
the process), earth or peat wax (such as ozokerite, ceresin, montan wax) or
natural waxes such as beeswax, camauba, or candelilla, can also be used.
So-called Japan waxes, ester waxes or high-molecular weight polyethylene
glycols are seldom used in hydrocarbon antifoams. Liquid hydrocarbons are
also used, even if only to a limited degree. Apart from mineral oils of na-
phthenic or paraffinic origin, vegetable or animal oils can be used. Common
to all these compounds is a low melting point of 40-50°C, with a minimum
Foam Control in Detergent Products 243

Sediment (ml)

-+- Silica A
1,2
--H- Silica B
-8- Silica C

0 10 20 30 40 50 60 70 80 90 100
Methanol in Water 1%)

FIG. 12 Methanol wettability of different types of hydrophobic silicas. With rising


content of methanol in water an increasing amount of silica is wetted which will
form a bigger sediment. The antifoam behavior of the silicas is shown in Fig. 13.

boiling point of l l0°C. The boiling point is important in that, at the max-
imum washing temperature of 95°C (boil wash), the antifoam components
must not be volatilized. Adverse effect due to odor and color mean that only
mineral oils such as spindle or paraffin oils have been able to achieve sig-
nificance in practice.
Very few hydrophobed silica/hydrocarbon antifoams are described in the
detergent patent literature (Table 6). Compared to this, waxes are widely
claimed in connection with silicone oil antifoams, either as a component of
the active system or as organic carrier. This will be discussed in detail in
connection with the silicone oil defoamers.
A common feature of the claimed antifoams listed in Table 6 is mixture
of a surface-active organic component with the hydrocarbon. For instance,
glycerol monostearate [74,75,79] in portions of 50-70%, or so-called Guer-
bet alcohols [76, 78], which are present in antifoams to an amount of 60-
80% are claimed. The function of these substances (ofHLB <00) is to assist
dispersion of the antifoam.
The antifoam system usually comprises a mixture of a wide variety of
hydrocarbons which, at most, has only few liquid constituents at ambient
temperature and which has components with melting points in some cases
just above 70°C. These mixtures are manufactured by mixing the individual
TABLE 6 Review of Patent Literature Dealing with Synthetic Silicas and Hydrocarbons (1978-1985)

Other organic Incorporationc


Silica• Hydrocarbonb ingredients Method Claimed advantages Reference
--
PS LH, SH, MW Fatty acid-ester I, 3, 5 Robustness 74
PS SH,MW Nonionic surfactant 2, 3, 4 Effective with complex 75
surfactant systems
FS, PS SH Guerbet alcohol 4, 5 Similar to silicon anti- 76
N
foams, broad application
range
"""" PS LH, SH 4, 5 Low amount of LH 77
FS, PS SH,MW Guerbet alcohol 2, 4, 5 Effective with high amounts 78
of anionic surfactant
PS LH, SH Fatty acid-ester 4, 5 Storage stability 79
'PS = precipitated silica; FS = fumed
silica
bLH = liquid hydrocarbon (mp < 20°C)
SH = solid hydrocarbon (mp > 30°C)
MW = microcrystalline wax
cl = spray drying; 2 = spray cooling; 3 = spraying on detergent powder; 4 = spraying on carrier; 5 = antifoam granules
Foam Control in Detergent Products 245

components as well as the emulsifier in the melt and subsequently incor-


porating the hydrophobic silica. In a first step, the silica is intermingled
using a blade or beam agitator, which is followed by the actual dispersion.
Dispersion is effected with high-shear mixers such as, for instance, Ultra-
Turrax or the Gaulin Homogenisator.
A variety of methods are used to incorporate the antifoam in detergent
powder. Here it should be noted that the method of incorporation has a
decisive influence on the extent to which antifoam activity is maintained and
the length of time the antifoam remains stable in the washing powder after
storage. In contrast with soaps, incorporation in the detergent slurry with
subsequent spray drying to produce washing powder is not recommended.
The hydrophobed silica/hydrocarbon antifoam would be inactivated during
preparation of the detergent slurry. The reasons for this are not well understood.
The following methods have been shown to provide more successful
incorporation:
Spraying onto the washing powder
Spray cooling
Spraying onto a carrier
Production of antifoam agglomerates
Spraying onto the powder is the simplest method. However, the dosing ac-
curacy must be high because relatively small quantities (approx. 0.1-1%)
must be uniformly sprayed onto the powder [74,75].
Direct spray-drying of the antifoam dispersion has also been suggested.
This is known as "spray cooling" because it is effected in spray towers with
cooled air [75,78]. It can only be employed with dispersions whose melting
point is 40°C and higher. As with direct spray-on, problems are encountered
with the homogeneous incorporation into the detergent powder of small
quantities of antifoam powder, particularly since the latter is difficult to han-
dle. It also has a high tendency to agglutinate on storage, so that antifoam
powders produced according to this method should be further processed
immediately.
From an applications engineering point of view, the manufacture of an-
tifoam agglomerates and the spraying of the antifoam dispersion onto car-
riers is advantageous. Preference is also given to these two methods in prac-
tically all the relevant patents. Antifoam agglomerates can be manufactured
by mixing the hot, finished antifoam dispersion in a kneader together with
salts which are preferably contained in the detergent. The simplest way is
to cool during mixing, the mass then being ground [76,77] or extruded [81].
Agglomeration of the powdery constituents with the dispersion is also pos-
sible, e.g., in a pan granulator.
246 Ferch and Leonhardt

The method of spraying onto a carrier requires separate preparation of


the carrier usually by spray-drying. This leads to highly absorptive granules
which are impregnated with the antifoam dispersion in a subsequent step.
Tripolyphosphate, carbonates, silicates, and sulfates can be used as carriers.
A carrier specially designed for this purpose consists of 5-15% water glass,
30-35% tripolyphosphate, 25-45% sodium sulfate, and 7-18% water [78].
Carriers of quite another type are gelatinized starch and perborate mono-
hydrate [77, 79]. These are both coarse-particle substances with a particle
size of 400-600 µm, and are extremely absorptive.
The advantage of the antifoam aggregates produced by these two methods
is that the antifoam is fixed in a surfactant-free, water-soluble or at least
easily dispersible matrix which means improved storage stability. Further-
more, the matrix serves to "dilute" the antifoam, so that instead of0.1-1%
antifoam up to 5% antifoam agglomerate is now to be added, which is a
distinct advantage as far as homogeneous and uniform incorporation in the
washing powder is concerned.
3. Hydrophobed Silica/Polydimethylsiloxane Antifoams
a. Properties. The most effective and most significant antifoams are the
so-called silicone antifoams. These are characterized by the fact that they
are not dependent on the water hardness and have a broad spectrum of ef-
ficiency as far as the most diverse surfactant systems and the most varied
application conditions are concerned. What distinguishes them from all other
antifoams, including hydrocarbon-based antifoams, is their high efficiency
at low dosages. Depending on the type of detergent, quantities of below
0.1 % can be sufficient to achieve a satisfactory foam profile in the washing
process. The majority of silicone oils used are polydimethylsiloxane poly-
mers which are also known as "silicone fluids." These linear polysiloxanes
are manufactured by selected hydrolysis of dimethyldichlorosilane and tri-
methylchlorosilane, the linear siloxane chains being formed during
condensation.

(2)
Me3SiO[-r--0]- SiMe3 + 2(n + l )HCl

Me n

The parameter n, given in the formula (2) for the degree of polymerization,
varies from about 20 to 2000, at times, to even higher values. This is also
Foam Control in Detergent Products 247

combined with a wide range of viscosities of from 50 cSt at n = 20 up to


200,000 cSt at n = 2000 [80-82]. The siloxane skeleton makes it possible
to densely pack the methyl groups which dominate intermolecular forces
and, consequently, determine the surface tension. The latter is relatively
insensitive ton for n > 20 and then has values of approximately 20-22 mN
m- 1 (at 20°C) [83]. This lies within the range of surface tensions of crys-
talline hydrocarbons, the crystal lattice of which forces dense packing of the
methyl groups [84]. Liquid paraffin hydrocarbons have surface tensions higher
by about 10 mN m- 1 than silicone oils.
Other physical-chemical characteristics of the silicone oils are l 83]
Clear, colorless liquids
Spontaneous dewetting and spreading onto the surface of the majority of
aqueous surfactant solutions
Insoluble in the aqueous medium
Incompatible with surfactants
Difficult to emulsify
Chemically inert
These characteristics make it easy to understand why silicone oils are ideal
antifoam liquids. Due to their extremely low surface tension, silicone oils
dewet and spread spontaneously over the surface of most surfactant solu-
tions. Thus the surface tensions of these solutions usually vary between 25
and 40 mN m -l at 20°C which clearly exceeds the air-oil surface tension of
a silicone oil [85]. The high efficiency of silicone antifoams can be derived
from this since the dewetting of antifoam into a foam film surface is a nec-
essary property (see Chapter 1).
In addition to the chain-form polydimethylsiloxane polymers, so-called
siloxanc resins have also achieved certain significance. These are three-di-
mensional, cross-linked polymers formed by Me 3Si0 112 and Si0 412 units.
The resins are manufactured from trimethylchlorosilane and silicone tetra-
chloride by means of cohydrolysis or condensation. Another variant is the
hydrolysis of alkyltrichlorosilanes. The resin usually has a portion of 0.5-
5% free silanol groups.
Silicone antifoams usually consist of mixtures of polydimethylsiloxane
and hydrophobed silica. The antifoam may be prepared by mixing hydro-
philic silica with polydimethylsiloxanc and subjecting the mixture to heat
treatment to ensure surface reaction between the silica and the polydimethyl-
siloxane to produce hydrophobed silica in situ. Alternatively silica hydro-
phobed by other methods may be directly mixed with the polydimethylsiloxane.
As with hydrocarbon-based antifoams manufacturing is divided in two
steps: preparation of antifoam dispersion and incorporation in the washing
powder. Much attention is paid to the latter in the patent literature. This is
248 Ferch and Leonhardt

illustrated by Table 7 where a list of silicone antifoam patents for detergent


powder application is presented. Practical experience has shown that the
method of incorporation of silicone antifoams has a decisive influence on
antifoam efficiency, particularly with reference to maintaining activity when
the washing powder is stored for a long period.
The first silicone antifoams with hydrophobic silica as promoter were
described at the beginning of the sixties [112, 113]. Here the silica was hy-
drophobed in situ by heating a silica/siloxane dispersion for several hours
to temperatures of 150-220°C, with a possible addition of Lewis acids such
as AlCh, FeC13 , SnCl4 , TiC14 , BF3 , ZnC1 2 as catalysts (112]. As we now
know, this heat treatment modifies the surface of the silica. The activated
silanol groups of the silica are present in a "silanate" form and can be re-
acted with the siloxane polymer in such a way that the basic silanate group
is deposited on the silicon atom of the siloxane polymer [114]. A penta
coordinated state of transition is passed through [ 115] and a new siloxane
unit forms during heating, the chain-form siloxane polymer being decom-
posed. As a result of this reaction between the silica and the siloxane poly-
mer, a silica with a hydrophobic surface is obtained.

Silicone oil RT
+ .,.. et-dispersion (3)
hydrophilic silica o. 5 hr
40-300°C
et-dispersion -----11111>- [3-dispersion (4)
1-10 hr
In a first stage, the silica is added to the silicone oil and a so-called "a-
dispersion" is produced under high-shear stirring conditions (3). Mixers which
allow chargewise production are available, for example the Ultra-Turrax.
The Dispax-Reaktor or Gaulin-Homogenisator are widely used for contin-
uous processes. The silica employed is either "activated" by treatment with
alkali hydroxides or the alkali or other catalysts functioning as Lewis acids
are added during dispersion [80,81,116-118]. Silicone oils with viscosities
below 5000 cSt are generally used at a ratio of from 85% to 97% to 3-15%
in the mixture with the silica. Siloxane resins are also sometimes used which
can react with the silica via the still freely available silanol groups (up to
5%) [81]. The dispersion is heated under inert gas for a specific period of
time to temperatures usually in the range of 150-200°C. Temperature and
time are dependent on the catalyst added. In the majority of the cases re-
action is completed after only 1-2 hr at temperatures below 200°C. The
resultant dispersion of a hydrophobic silica in silicone oil is also known as
a "13-dispersion" (4).
Foam Control in Detergent Products 249

An example is given in Fig. 13. Two precipitated and one fumed silica
were hydrophobed in situ with a 1000-cSt silicone fluid (DC 200, Dow
Coming). The silica portion of the antifoam was 5%. The silica was acti-
vated to the highest degree by soda lye to give a high pH and was conse-
quently converted with the silicone oil. A measurement for this is the meth-
anol wettability. The curves of the methanol wettability of the three silicas
are given in Fig. 12 indicating that silica A is the most "hydrophobed" one.
A foam test in a washing machine performed to determine antifoam effi-
ciency showed silica A to be the most effective (Fig. 13).
In situ hydrophobing has the advantage of high flexibility, i.e., adaptation
to the individual requirements. However, it is a relatively expensive process.
It is simpler and usually more cost favorable to manufacture the antifoam
directly from a hydrophobic silica and the silicone oil. Special antifoam
silicas which are described above are available for this purpose. If different
silicas are examined for their initial (immediately after mixing with poly-
dimethylsiloxane) antifoam efficiency it becomes apparent that degree of
hydrophobicity primarily determines antifoam efficiency. This is illustrated
in Fig. 14 where it is clear that antifoam efficiency correlates with hydro-
phobicity as indicated by methanol wettability. As we have indicated the
method of incorporation of silicone antifoams in detergent powders is of
critical importance. This is a consequence of the tendency of silicone an-
tifoams to lose their effectiveness when mixed with detergent powders.
Deactivation processes are particularly noticeable with very low levels of
antifoam. Usually, the antifoam is added to the detergent powder in the form
of a 10-50% antifoam adjunct. The antifoam can diffuse from these con-
centrates into the powder matrix. This process has been suggested by Saw-
icki (83] as a possible mechanism to explain deactivation.
Here the loss of effectiveness is explained by a change in the droplet size
of the antifoam as a result of such diffusion. Thus diffusion into the powder
matrix supposedly results in release of "smaller" antifoam droplets when the
powder is dispersed in the wash liquor. Such droplets are supposed by Saw-
icki to be less effective than the "larger" droplets released before diffusion
has taken place. No direct evidence for this hypothesis is however given
where measurement of the relevant droplet size has been made.
On the whole then there would appear to be no clear explanation for the
deactivation of silicone antifoams when in contact with detergent powder.
b. Incorporation Methods. In essence, three methods have been devel-
oped to prevent deactivation, which can be subdivided into a number of
variants. A compilation is given in Table 8. These methods are

Embedding the defoamer in an organic matrix, preferably of substances with


a melting point above 35°C
Table 7 Review of Patent Literature Dealing with Synthetic Silicas and Silicone Oils ( 1971-1988)
Method ofe
Silicasa Polysiloxaneb Hydrocarbon< Other ingredientct processing Claimed advantages Reference

FS", PS" LVS - - Cl Granulated antifoam 82


FS" LVS/SR - PEG 4,000, A2, A3 Storage stability, insen- 86
TA25EO sitivity toward water
hardness
FS" LVS/SR MW PEG 4,000, A2,A3 Low dosage, synergism: 87
TA25EO silicone carbonwax
FS", PS" LVS MW Nonionic, CMC, PVA B3 Good powder dispers- 88
ability, homoge-
neously distributed
anti foam
-
N
U1
Q FS", PS" LVS Nonionic, CMC B2 Storage stability, no 89
fabric graying
FS", PS" LVS/SR - C2 Good powder dispers- 90
ability, no activity
loss during processing
FS", PS" LVS/SR - - C2 Good powder <lispers- 91
ability, no graying of
fabrics
FS~ LVS SH/MW Succrose, Ti0 2 , CMC Cl Storage stability 92
FS", PS" LVS - CMC, MC, PVA B2 Storage stability, homo- 93
geneously distributed
anti foam
FS" LVS/HVS.SR - Nonionic, siloxane- Al, A3, A4 Storage stability, good 94
oxyalkylene dispensing properties
polymer
FS", PS" LVS SH Gelatinized starch Cl Improved storage 95
stability
FS 13 LVS MW Nonionic, PEG A3 Mineral oil/silicone oil 96
antifoam effective at
high surfactant levels
FS" LVS/HVS SH/MW Diamide. alkylphos- Cl No organic solvent for 97
phate, gelatinized processing of the
starch hydrocarbons
FS" HVS - Nonionic, CMC BI Storage stability 98
FS" LVS/HVS SH/MW CMC/MC B3 Storage stability, no 99
surfactants in
anti foam
FS" LVS/HVS - Siloxane-oxyalky- A3 High robustness in corn- 100
lene-copolymer bination with 0.5-

...
1.5% soap
JOI
N
U1
PS" LVS - Siloxane-oxyalky- A4 Softening effect remains
lene-copo lymer, uneffected
TAllEO
FS" LVS/SR PEG, soap, TA25EO A3 Storage stability, good 102
antifoam properties
psl3 LVS SH/MW Gelatinized starch, Cl Storage stability, low- 103
perborate, zeolite, and high-temperature
salts anti foam
psl3 LVS LH Gelatinized starch, Cl Storage stability 104
diamide,
alkyl phosphate
FS 13 , PS~ LVS/HVS SH CMC, MC, starch A3 Storage stability, simple 105
process
FS~, PS 13 HVS - Fatty acid alcohol A2,A3 Storage stability, simple 106
anti foam
Table 7 Review of Patent Literature Dealing with Synthetic Silicas and Silicone Oils ( 1971-1988)

Method ofe
Silicas• Polysiloxaneb Hydrocarbon< Other ingredientd processing Claimed advantages Reference

FS 11 , PS 11 HVS - Glycerol-monoester A2,A3 Storage stability 107


ps11 LVS/HVS LH/SH/MW Fatty acid, soap, A2,A3 Storage stability be- 108
diamide, cause of soap matrix
alkylphosphate
psll LVS/HVS LH/SH/MW Special soda ash C2 Porous carrier, storage 109
.., -
stability
110
..,
UI FS11 LVS PEG 5,000-10,000,
TA25EO
A3 Storage stability, pro-
tective envelope
against surfactants
FS 0 , PS 0 HVS - Nonionic, CMC, B3 Good foam pattern be- 111
PA, PVA, PVP cause of combination
silicone-anionics

"FS = fumed silica, PS = precipitated silica; a = hydrophobic and hydrophilic, J3 = only hydrophobic.
bLVS = low-viscosity siloxane <5000 cSt; HVS = high-viscosity siloxane > 5000 cSt; SR = silicone resin.
CLH = liquid hydrocarbon (mp < 20°C), SH = solid hydrocarbon (mp > 30°C); MW = microcrystalline wax.
•PEG= polyethyleneglycol; TA= tallow alcohol; CMC = carboxymethylcellulose; MC= methylcellulose; PA= polyacrylate; PVA = polyvinylalcohol,
PVP = polyvinylpyrrolidone.
"Differentiation according to the compilation given in Table 8.
Foam Control in Detergent Products 253

SILICA A B c
Type preci- preci- fumed
pitated pitated
BET-surface !m2/g] 170 190 200
Mean agglome - [µm] 5 7
rate size
pH-value 8.3 6.3 4.0

+""=----r---.----r-----,-----,-----1. 0
10 20 30 40 50 60
Time (min)

FIG. 13 Primary antifoam behavior of different hydrophilic silicas (A = FK 383DS,


B = Sipemat 22S, C = Aerosil 200, all from Degussa) dispersed by 5% in silicone
oil of 1OOO cSt (DC 200, Dow Coming). After heating for 5 hr to 220°C and a final
homogenization 10 parts of the dispersion were mixed with 90 parts granulated zeo-
lite A (Wessalith CS, Degussa) thus forming an antifoam granulate. 2% of this
granulate were mixed to a STPP-built detergent (8% linear alkylbenzene sulfonate,
4% nonionic surfactant, 16% zeolite A, 22% STPP) and immediately tested.

Microincapsulation using film-forming polymers


Impregnation of absorptive detergent constituents or specially designed carriers

The objective is the same with all three processes: immobilization of the
antifoam by fixing it to a carrier and protection from the detergent constit-
uents and the moisture in the powder. In this way, the antifoam is simul-
taneously "diluted" so that low feed quantities can be homogeneously dis-
254 Ferch and Leonhardt

SILICA D E F
Type preci- preci- fumed
pitated pitated
BET-surface [m2/g] 90 100 100
Mean agglome - [µm] 5 10
rate size
pH-value 10.3 8.0 4.0
Hydrophobicity [%] »50 <50 «50
(MeOH-wettability)

0 +----.----+----.,.----,---~------'- 0
0 10 20 30 40 50 60
Time (min)

FIG. 14 Primary antifoam behavior of different hydrophobed silicas (D = Sipemat


DIO, E = Sipernat Dl7, F = Aerosil R 972, all from Degussa) dispersed in silicone
oil of 1000 cSt (DC 200, Dow Corning). Measurement is carried out as described
in Fig. 13.

tributed in the washing powder. Several important patents will be discussed


in detail as illustrative of the processes referred to above (see Table 7).
Use of organic matrices. The first description of a silicone antifoam
embedded in an organic matrix is almost 20 years old [86]. The antifoam
system comprises 10-50% fumed silica and 50-90% polydimethylsiloxane
(PDMS). The silica is characterized by a BET surface of at least 50 m2 /g,
a primary particle size of 10-20 nm and can be either hydrophilic or hy-
drophobic (silanized). A dispersion is produced from this and PDMS which
can have a molecular weight of between 200 and 200,000. The antifoam is
Foam Control in Detergent Products 255

TABLE 8 Compilation of Different ways of Manufacturing Silicone-Based Antifoams

Processing Type

Embedding the antifoam in an organic matrix


spray-drying with salts Al
spray-cooling A2
granulation A3
spraying on detergent powder A4
Microincapsulation using film-forming polymers
precipitation out of an aqueous emulsion BI
spray-drying B2
granulation B3
Impragnation of a carrier
granulation and coating Cl
special carrier for impragnation C2

in tum dispersed in a water-soluble or water-dispersible non-surface-active


carrier. Examples of such carriers include polyethylene glycol, with a mo-
lecular weight of 4000 (PEG 4000) or a tallow alcohol with 25 mol ethylene
oxide (TA25EO). Here it is supposed that due to the degree of ethoxylation
the surface-active characteristics of the TA25EO are very weak. About one-
third part of the system is silicone oil/silica dispersion and the remainder
consists of the organic carrier. This mixture may then either be spray-cooled
or be sprayed onto detergent constituents, such as tripolyphosphate, in a
fluidized-bed mixer. In the latter so-called "coated agglomerates" develop.
A typical agglomerate has for example the composition 5% silica, 5% sil-
icone fluid, 40% TA25EO, and 50% tripolyphosphate. The agglomerate size
is between 250 and 1500 µm.
The antifoam agglomerates produced in this manner are characterized by
high storage stability in detergent powder. Good foam regulation over the
entire temperature range, distinctly reduced graying of the washing and no
dependence on the water hardness are said to be the advantages. Dosage of
the antifoam agglomerate is adjusted in such a way that the content in the
detergent powder of active antifoam is between 0.01 % and 0.S'fo.
This antifoam system can be further developed by combining it with waxes
(87). Synergistic effects are claimed, which allow the feed concentrations
of the wax and silicone to be drastically reduced. The wax component is a
microcrystalline petroleum wax with a melting point of 77°C. The required
dosage is up to 2%, whereas the dosage of the silicone antifoam can be
reduced to between 0.04% and 0.1 % . The ratio of wax to silicone antifoam
can be varied within the limits 20: I to 1: 10. The silicone antifoam corn-
256 Ferch and Leonhardt

ponent is, as described above, either produced by spray cooling or by ag-


glomeration with sodium tripolyphosphate in a fluidized bed. The wax is
dissolved separately in the nonionic surfactant of the detergent (which con-
tains 12% of a primary C 14_ 15 alcohol with 7 mol ethylene oxide). This mix-
ture is then added to the detergent slurry before spray-drying. The combi-
nation of silicone antifoam and wax is said to have a foam regulation effect
both in the wash cycle and in the rinsing cycle in the case of nonionic sur-
factant-based detergents.
Along with pure polydimethylsiloxane polymers, modified siloxane poly-
mers, so-called siloxane oxyalkylene alkoxysiloxane copolymers, are also
used [94,96,100,101]. These compounds are derived from PDMS in which
oxyalkyl groups are substituted for the methyl groups (§).

p + r = 30-120; p:r = 2:1-8:1 (§)

The radical R is claimed to have the following composition:


R = -X-(EOb;2-POb;2)0R,
X = --CH2 ,--C2H4 ; b = 10-30; R, = -H,--CH3 ;
EO =ethylene oxide, PO= propyleneoxide
Alkoxysiloxanes have the advantage that they can improve the stability of
the silicone antifoam dispersion in the organic carrier if the organic carrier
is a nonionic surfactant with a portion of EO groups [ 100]. This stabilizing
effect results in the antifoam droplets being retained in the carrier matrix
and ensures effective storage stable incorporation in the detergent powder.
A typical antifoam system with alkoxysiloxane as emulsifier has the follow-
ing composition:
Polydimethylsiloxane (PDMS) 17.5-23.8%
Hydrophobed precipitated silica 6.0-7.9%
Alkoxysiloxane 1.7-2.0%
TAllEO 74.8-66.3%
It is manufactured by mixing the nonionic surfactants with alkoxysiloxane
and then high-shear conditions. The resultant emulsion can be directly sprayed
onto the washing powder (dosage 0.1-1.5% [101]) or agglomerated with
other detergent constituents in the fluidized bed [94]. The advantages offered
Foam Control in Detergent Products 257

by this method of manufacturing antifoam systems are said to be high stor-


age stability and good dispensing properties of the powder. A negative in-
fluence of PDMS on the wetting behavior of detergent powders is apparently
avoided.
A system which is particularly suitable for detergents containing more
than 12% surfactants is manufactured according to a multistage process [100].
Hydrophobic fumed silica (1-25%) are stirred into PDMS (75-99 %) having
viscosity of 20-12,500 cSt and dispersed by means of an in-line high-shear
pump (e.g., Dispax from Janke and Kunkel). When the temperature has
risen to above 95°C and the viscosity to 6-10,000 cSt, high-viscosity PDMS
(>25,000 cSt) in a portion of 33% to 66% is added, and dispersing is con-
tinued. The advantage of the high-viscosity PDMS is to avoid or, at least,
to minimize deficiency for foam control in high active detergent composi-
tions, presumably because the PDMS is rapidly dispersed or solubilized by
the high surfactant level. This premix is added to a mixture of organic carrier
and alkoxysiloxane and homogenized under high-shear conditions. The or-
ganic carrier is a twofold one consisting of a highly ethoxylated nonionic
surfactant (TA25EO, TA80EO) or polyethylene glycol (PEG 6000, mp 38-
900C) and a low-ethoxylated alcohol (e.g., C8_ 24 x 2-12 EO, mp 5-36°C)
at a ratio of 2-25% to 57-98%, in which 0.5-8% of the alkoxysilicone is
contained. The ratio of the premix to carrier is approximately 1: 1'6-25 when
the antifoam dispersion is directly sprayed onto the washing powder (de-
tergent dosage 0.25-0.65%), or 1 : 1-2, if the antifoam is to be agglom-
erated (detergent dosage 0.4-0.65%). The agglomerates consist to 50-85%
of an inorganic carrier such as tripolyphosphate or other salts and the organic
carrier /premix mixture. Pan and fluidized bed agglomeration or agglom-
eration according to the so-called Schugi process can be used to produce the
agglomerates. This antifoam system avoids the disadvantages of other sys-
tems, such as the incompatibility of high PDMS/silica quantities in the car-
rier matrix based on nonionic surfactants and the too slow kinetics during
transportation of the antifoam to the air/water interface in the wax/silica
dispersion system.
A special matrix, called "acid soap," consists of a mixture of 40-60%
lauryl acid and the residual sodium stearate or sodium palmitate having a
liquid portion below 1% at 40°C [108]. In addition, this carrier has certain
foam-inhibiting properties. The antifoam itself can be either a hydrocarbon-
based antifoam or a PDMS /hydrophobed silica mixture. It is present at 40-
45% in the matrix. The antifoam system may be manufactured by spray-
cooling or by cooling the hot mixture on a band drier and subsequently
grinding the flakes. Extrusion to needles, such as is known from manufac-
turing soap needles, is also suggested. For reasons of stability it is important
that the particle size of the needles or the ground flakes is above 500 µm.
258 Ferch and Leonhardt

Glycerol monoesters like glycol monostearate [ l 07] or fatty acid/fatty


alcohol mixtures of compounds with 12-20 carbon atoms, such as stearyl
acid and alcohol ! 106 J, are also suggested as organic matrices. Anti foam
adjuncts suitable for incorporation in detergent powders are manufactured
by spray cooling or in the fluidized bed with tripolyphosphate salt.
Special manufacturing processes are roller compacting [ 105] and the
production of flakes on a belt cooler [102, 110]. Roller compacting is a
process in which a relatively dry powder mixture is forced under pressure
(30-100 bar) through a slit with an adjustable gap width. The compacted
strip produced is ground and graded into the desired particle size via
screens. The porosity of the agglomerate produced according to this pro-
cess is very low which contributes to the stability of the antifoam. The ini-
tial powder mixture consists of a base powder, e.g., sodium sulfate, onto
which the antifoam is sprayed. A little water can also be sprayed on to in-
crease plastication, and layered silicates or zeolite can be added to improve
redispersibility.
Flakes are produced on a belt cooler, which can have a thin film of water
to improve heat transfer [ 110 I. The organic matrix comprises more than 80%
of polyethyleneglycol of molecular weight of 5000-10,000 (mp 43°C), which
may contain small amounts of soap (0.2-15%) [102] or TA25EO (0.25-
2.00%) [110]. Both additives serve to retard release of the silicone antifoam.
The antifoam is made up of I 00 parts PDMS, 5-20 parts siloxane resin,
and 1-5 parts of a hydrophobic fumed silica. A dispersion is prepared from
the antifoam constituents which is incorporated into the molten carrier.
For reasons of efficiency homogenization should be carried out in such a
way that the antifoam droplets have a size of 5-30 µm before the disper-
sion is cooled to flakes. The ground flakes are added to the detergent at
a ratio of 2.5-10% depending on the antifoam content, so that the effec-
tive antifoam level is 0.01-0.5%. The antifoam has an improved storage
stability and the matrix is said to prevent migration of the antifoam in the
powder.
Use of film-forming polymers. A quite different method of incorporating
silicone antifoams in detergent powders is to use film-forming polymers.
The term film-forming polymers refers to cellulose ethers, such as carboxy
methyl or methyl cellulose, polyacrylate, polyvinylacetate (PY A), polyvinyl
alcohol, polyvinyl pyrrolidone (PVP), and similar compounds. Here the sil-
icone antifoam is dispersed as an oil-in-water emulsion in an aqueous so-
lution or dispersion of polymer. The purpose of the polymer is to form a
barrier layer between the antifoam dispersion and the washing powder. The
film-forming polymer wraps itself protectively around the antifoam which
will form a microcapsule or an agglomerate coated with the washing powder
Foam Control in Detergent Products 259

is delayed to a high degree under storage conditions which explains the sta-
bilizing effect.
A particularly positive effect on the stability of silicone antifoam disper-
sions is reported when cellulose ethers are used. This is accompanied
by simplification of the manufacture of the relevant antifoam emulsions.
Thus it is possible to operate without using emulsifiers which is consid-
ered in many cases as being of advantage from the point of view of storage
stability.
Microcapsules containing the antifoam dispersion can be produced by
manufacturing a silicone oil/silica emulsion using nonionic surfactants, add-
ing this to an aqueous solution of the sodium salt of the carboxymethylcel-
lulose and precipitating the cellulose, simultaneously destabilizing the emul-
sion [98]. Precipitation is performed with multivalent cations. Aluminum
salts, for example, have proved to be particularly useful. Hydrosoluble al-
cohols can also be used as precipitants. After precipitation washing and drying
in a vacuum powders with an antifoam content of more than 50% are ob-
tained. The resulting antifoam adjunct can be easily metered into a washing
powder and has a very good storage stability.
A somewhat different variant includes spray drying [93]. The film-form-
ing polymers (cellulose, ethers, starch ethers, polyacrylates, PY A, PVP) are
allowed to swell and dissolve; the polymer concentration is 1-5%. A PDMS/
silica dispersion is added (approx. 5-20%) and homogenized under high-
shear conditions. The dispersion droplets should have a size of about 1-20
µm. The polymer is precipitated with 20-50% electrolytic salts, such as
chlorides, sulfates, nitrates, silicates, and phosphates, to form microcap-
sules. This microcapsule suspension is conveyed separately from the deter-
gent slurry to the spray tower. It is combined with the detergent slurry just
before the spraying nozzle. The microcapsules are retained and are homo-
geneously distributed in the washing powder. High storage stability and the
relative absence of weight-controlled process steps are the stated advantages.
The dosage level is between 0.1% and 0.15%.
With the aid of cellulose ethers direct spray-drying of a silicone oil/hy-
drophobed silica-dispersion with inorganic salts and small quantities of non-
ionic surfactants is also possible [891. Two percent to 8% of a PDMS / Acro-
sil dispersion (90-98%:2-10%), 3-8% nonionic surfactant (C 12-C 20 with
4 to 25 ethyleneoxide groups), 3-8% cellulose ether, and 75-90% salt (pref-
erably sodium sulfatc) arc homogenized and spray-dried. If the mixture con-
tains more than 50% sodium sulfate, then the water required for sodium
sulfate decahydrate formation is sufficient for the slurry, which has tem-
peratures above the melting point of sodium sulfate decahydrate of 33.4°C,
so that when spray drying, in principle, the cooling process for agglomerate
formation is adequate and no water need be dried off. A typical composition
260 Ferch and Leonhardt

is as follows:

PDMS 6.0%
Aerosil, silanized 1.5%
TA14EO 5.0%
TA5EO 3.0%
Na-CMC 7.0%
Na-silicate 10.0%
Na-sulfate (H 20 free) 64.5%
Water 1.5%
In addition to the good storage stability special emphasis is placed on an-
tifoam efficiency, avoidance of fabric graying, and prevention of siloxane
film formation in the machine.
A manufacturing process already known in connection with embedding
of the antifoam in an organic carrier is agglomeration [88, 99, 111]. As a
rule, an emulsion is produced from the silicone antifoam, the film-forming
polymers, nonionic surfactant, or ester waxes, with which the carrier salts,
such as tripolyphosphate, Na-sulfate, zeolite, and Ca-carbonate (soda ash)
are agglomerated. Agglomeration can be performed in simple drum, tum-
bler, or plough-blade mixers (Lodige).
The use of hydrosoluble polymers in combination with waxes is consid-
ered advantageous, since this improves the homogeneity of the distribution
of the PDMS/silica antifoam on the carrier [88]. A storage-stable antifoam
agglomerate is produced by agglomerating 68.8 parts sodium sulfate and
10% magnesium silicate with a dispersion of 5.6% PDMS (13,000 cSt),
0.4% Aerosil, 6% TA14EO, 0.2% CMC, and 9% water in a rotary drum
granulator. Seven percent of this agglomerate containing 4% silicon anti-
foam is added to a washing powder, which itself consists of two agglom-
erates, a spray-dried base powder and a sodium perborate tetrahydrate im-
pregnated with a nonionic surfactant. The separation of a detergent into three
individual agglomerates results in improved wetting and dispersing behavior
of the detergent powder.
A special antifoam system is a combination comprising a high-foam an-
ionic surfactant (e.g., alkylsulfate, alkylether sulfate, a-sulfosuccinoester)
and a silicone antifoam [ 111]. This system is deposited on a carrier, pref-
erably of sodium carbonate (soda ash), and cellulose ether, in a Lodige mixer
and encapsulated with the film-fonning polymer. What is important in this
case is the use of the anionic surfactant as powder and the sequence in which
the substances are mixed together. The sequence should be inorganic salts
and cellulose ether, powdery surfactants, silicone antifoam and, last, the
film-forming polymer. Among the film-forming polymers copolymers of vi-
Foam Control in Detergent Products 261

nylacetate and short-chain carbonic acids, such as crotonic acid, have proved
effective. About 0.2-2.5% of the antifoam agglomerate is used; with 10%
silicone antifoam in the agglomerate, this means a dosage level of 0.02 to
0.25%. The bulk density of agglomerate is around 550-600 g/L, it has a
particle size distribution similar to that of a detergent, has a good granular
hardness and is free flowing.
Use of absorptive carriers. The last of the three manufacturing processes
concerns the impregnation of a carrier with a silicone oil/hydrophobed silica
dispersion. Here agglomeration processes use mechanical mixers such as
Eirich pan granulators, Schugi mixers, fluid-bed mixers, drum and plough-
blade mixers (Lodige). In the simplest case an antifoam agglomerate is pro-
duced by impregnating 60 parts of an anhydrous tripolyphosphate with 40
parts of a dispersion prepared from 70% silicone oil antifoam (PDMS: hy-
drophilic fumed silica = 95:5) and 30 parts water [82]. The amount of water
is calculated in such a way that it is sufficient for complete hydration of the
tripolyphosphate; subsequent drying is not necessary. With a 2% dosage the
detergent contains 0.53% silicone oil and 0.028% silica. However, the stor-
age stability of this antifoam is still not satisfactory.
A carrier of a quite special nature consists of an inert core material of
spray-dried sucrose (25-80%) which is coated with an absorption agent ( 15-
40%) such as starch, titanium dioxide, CMC, silica, or calcite [92]. About
5-30% of a PDMS /hydrophobic silica dispersion diluted with an organic
solvent is sprayed onto a dry premix of core material and absorption agent
in an Eirich pan granulator. After evaporation of the solvent the agglomerate
is coated (3-30%) with paraffin (mp 35-65°C) or microwax (mp 60-63°C)
dissolved in the same solvent. In this case the wax serves exclusively as a
protective envelope for the antifoam agglomerate. A typical composition is
60.2% sucrose beads, 24% starch, 12% silicone antifoam, and 4.8% paraffin
wax. The dosage in the washing powder is between 0.3 and 5%. The regular
form of the core material is stressed as being of particular advantage since
the antifoam can be more easily absorbed on this, thus making coating sim-
pler. On the whole, this results in a storage stable antifoam system.
Gelatinized starch has been suggested as carrier for silicone antifoam
[95, 97, 103, 104]. This starch type is produced from a suspension which con-
tains the starch particles and which is dried in a steam-heated drum to a
partially hydrolized starch. It was found that the nonionic character of the
starch and the relatively low pH range of 5.5-9 have a stabilizing influence
on the antifoam. The small particle size of 50 to 500 µm provides an ad-
equately large adsorption surface while the narrow particle size distribution
is responsible for uniform and regular agglomerates. Particularly good stor-
age stability can be achieved if the agglomerate is coated with a wax whose
262 Ferch and Leonhardt

melting point is appropriate to the purpose [951, or a relatively high melting


nonionic surfactant [97), such as, for instance, C 8-C 20 with 5-20 moles
ethylene oxide. An antifoam system is, for example, composed of 45-55%
gelatinized starch, 35-50% of a silicone oil/silica dispersion, and 7-15%
paraffin wax. It is manufactured according to well-known processes, e.g.,
pan granulators, Schugi mixers, or in a fluid bed. First, the starch is ag-
glomerated with the silicone antifoam and subsequently coated with a so-
lution of wax in an organic solvent or with molten wax. The wax can be
more easily and uniformly applied as a solution in a solvent, however, this
process is extremely problematical in respect of industrial-scale production
since the solvent must evaporate. A variation of this antifoam system is to
use two antifoams, silicone oil/silica, and hydrocarbon/silica, or hydro-
carbon/alkylphosphate or hydrocarbon/diamide [103,104]. The silicone an-
tifoam is referred to as "a low-temperature-sensitive antifoam" because it is
a viscous fluid dispersion at room temperature, whereas the antifoam, due
to its ongoing melting range of 50-80°C, is characterized as "high-temper-
ature-sensitive antifoam."
Special attention is drawn to freedom from surfactants and high absorp-
tivity with respect to the development of absorptive carriers. These carriers
may be produced by spray-drying. This has been found to have a positive
influence on the absorptivity. For example, a carrier produced from 15% to
65% tripolyphosphate, 15-65% sodium sulfate, 5-15% water glass, and the
remainder water has a particle size of 200-800 µm with a bulk density of
500-650 g/L [90,91]. About 4-12% antifoam dispersion, comprising 86-
96% silicone oil or a silicone oil/siloxane resin mixture and 4-14% silica,
can be deposited on these carriers. A special carrier with a highly porous
structure is based on spray-dried carbonates or sulfates (modified Burkeit)
to which polymers are added [ 119]. This structure is characterized by a pore
volume of 0. 2- l cm3 / g, an average pore size of < 20 µm, and an average
particle size of 80-2000 µm. The absorptivity is stated to be very good and
the small pore size is said to have the advantage that small antifoam droplets
are released which promotes rapid transportation of the antifoam to the air-
water interface. The effectiveness of the antifoam is retained during storage
in the detergent powder until it is needed at the point of use. The antifoam
is either a mixture of silicone and hydrophobed silica or a hydrocarbon com-
bined with a so-called antifoam promoter such as silica, alkylphosphates, or
diamides [ 109]. Mixtures of the silicone and hydrocarbon-based antifoams
may be used. The antifoam systems are sprayed onto the carrier one after
the other. It is first impregnated with the silicone antifoam and then coated
with the hydrocarbon antifoam. Nonionic surfactants may be incorporated
to prevent the formation of "oil spots" on fabric surfaces derived from hy-
drophobic antifoam oils.
Foam Control in Detergent Products 263

V. CONCLUSION
The antifoams currently used in detergents represent the final point, for the
time being, of a development which was characterized by two points of
departure. On the one hand, the consumer demanded a detergent which ful-
filled a wide range of requirements. Namely, for example, an improved soil
removal capacity, better and milder removal of bleachable stains, applica-
bility for the most varied fibers and types of fabric under widely differing
application conditions (hand wash, all-temperature wash, soft-water condi-
tions). On the other hand experiments were focused on attempts to better
understand the physical-chemical correlations involved in the formation of
foams and their stability.
Development of the antifoams used in detergents today has only been
possible on the basis of the knowledge gained with respect to the antifoam
effect of hydrophobic particulate materials. The concept of hydrophobic an-
tifoam liquids (mineral oil and silicone oil) combined with hydrophobic par-
ticles has made it possible to develop numerous antifoam systems which can
be adapted to any respective "foam problem." One focal point of future
development will probably still be the formulation sphere. Further devel-
opments can be expected in this sector, retention of activity with long-term
storage being the driving force.

REFERENCES
l. J. Falbe, Surfactants in Consumer products, Theory, Technology and Ap-
plication, Springer, Heidelberg, 1987, p. 208.
2. G. Jakobi and A. Lohr, Detergents and Textile Washing, Principles and Prac-
tice, VCH, Weinheim, 1987, pp. 206
3. J. Falbe, Surfactants in Consumer products, Theory, Technologie and Ap-
plications, Springer, Heidelberg, 1987, p. 24 l.
4. P. L. Lagman, Chem. Eng. News, Jan. 23: 31 (1984).
5. H. Stache, Tensidtaschenbuch, Hanser, Miinchen, Wien, 1984, pp. 253-294.
6. J. Falbe, Surfactants in Consumer Products, Theory, Technologie and Ap-
plication, Springer, Heidelberg, 1987, pp. 223-225.
7. E. Schmadel, Fette Seifen Anstrichmittel 70: 491 (1968).
8. J.C. Harris, Oil and Soap 23: IOI (1946).
9. H. L. Peper, Coll. Sci. 13: 199 (1958).
JO. W. D. Harkins and E. K. Fischer, J. Chem. Phys. 1: 852 (1933).
11. E. Schmadel and C. P. Kurzendorfer, Waschmittelchemie (Henkel KGaA,
ed.), Hiithig, Heidelberg, 1976, p.122.
12. ibid., p. 104.
13. W. L. St. John and W. J. Griebstein (assigned to Procter & Gamble), DE
1,056,316; January 27, 1966; filed October 27, 1956.
264 Ferch and Leonhardt

14. P. Krings, G. Jakobi, and J. Galinke (assigned to Henkel & Cie GmbH) DE-
OS 2,243,306; March 21, 1974; filed September 2, 1972.
15. W. Leonhardt, unpublished results.
16. A. Sagredos (assigned to Unilever N.V.) DE 1,617,216; April 1, 1967; filed
February 24, 1967.
17. K. Bott, H. Grossmann, and K. Kosswig (assigned to Chemische Werke Hills
AG), DE-OS 2,244,665; March 21, 1974; filed September 12, 1972.
18. 0. F. Schwiegel, W. Griess, G. Ulrich, and M. Knausenberger (assigned to
Unilever N.V.), DE 1,617,227; June 6, 1974; filed June 12, 1967.
19. No inventor named, (assigned to Thomas Hedley & Co. Ltd.), GB 808,945;
February 11, 1959; filed May 16, 1957.
20. E. A. Schwoeppe (assigned to Procter & Gamble), DE 1,080,250; April 21,
1960; filed May 21, 1957.
21. H. Y. Lew (assigned to Chevron Research Co.), US 3,231,508; January 25,
1966; filed March 18, 1964.
22. H. Y. Lew (assigned to Chevron Research Co.), US 3,285,856; November
15, 1966; filed March 18, 1964.
23. H.J. Stimberg, J. Galinke and E. Schmadel (assigned to Henkel & Cie GmbH),
DE-OS 2,043,087; March 2, 1972; filed August 31, 1970.
24. P. R. Garrett (assigned to Unilever PLC), EP 75,433; May 28, 1986; filed
September 14, 1982.
25. M. N. A. Carter, P. R. Garrett, D. Giles, and A. R. Naik (assigned to Un-
ilever PLC), EP 87,233; November 12, 1986; filed February 3, 1983.
26. G. C. Vandenbrom (assigned to Unilever N.V.), EP 126,500; November 28,
1984; filed April 11, 1984.
27. P. Schulz, J. Harer, C.-P. Kurzendorfer, F.-J. Carduck, F. W. Diekotter,
U. Jahnke, and E. Schmadel (assigned to Henkel KGaA), EP 309,931; April
5, 1988; filed September 23, 1988.
28. E. Schmadel, Fette Seifen Anstrichmittel 70:491 (1968).
29. E. Gotte and E. Schmadel (assigned to Henkel & Cie GmbH), DE 1,257 ,338;
July 25, 1968; filed February 11, 1965.
30. E. Schmadel (assigned to Henkel & Cie GmbH), DE 1,467 ,620; August 6,
1970; filed December 15, 1965.
31. H.-J. Lehmann and E. Schmadel (assigned to Henkel & Cie GmbH), DE
1,617,116; October 16, 1975; filed June 25, 1966.
32. M. Berg and E. Schmadel (assigned to Henkel & Cie GmbH), DE l ,617, 127;
February 18, 1971; filed April 1, 1967.
33. G. Amberg and H. Saran (assigned to Henkel & Cie GmbH), DE-OS 2,333,568;
January 30, 1975; filed July 2, 1973.
34. J. Glas!, H. Saran, J. Hoffmeister, and M. Berg, (assigned to Henkel & Cie
GmbH), DE-OS 2,431,581; January 22, 1976; filed July l, 1974.
35. G. Amberg (assigned to Henkel & Cie GmbH), DE-OS 2,544,034; April 7,
1977; filed March 10, 1977.
36. J. Perner, G. Frey, and H. Helfert (assigned to BASF AG), DE 2,710,355;
August 10, 1978; filed March 10, 1977.
Foam Control in Detergent Products 265

37. J. T. Thurston, J. Am. Chem. Soc. 73: 2981 (1951).


38. D. W. Kaiser, J. Am. Chem. Soc. 73: 2984 (1951).
39. J. R. Campbell, and R. E. Hatton, J. Org. Chem. 26: 2786 (1961).
40. E. Jungermann and H. C. Silberman, in Anionic Surfactants, Part II (W. H.
Linfield, ed.), Marcel Dekker, New York, 1976, chap. 15.
41. H. Schlecht, H. Distler, and D. Stockigzt (assigned to BASF AG), DE-OS
1,362,570; June 26, 1975; filed December 17, 1973.
42. D. W. Farren and C. W. Stuttard (assigned to Unilever N.V.), DE-OS
2,537 ,570; March 11, 1976; filed May 9, 1974.
43. H. Distler and P. Diessel (assigned to BASF AG), DE 2,727,382; January
4, 1979; filed June 18, 1977.
44. M. D. Key and W. G. McNee (assigned to Lever Brothers Co.), US 4,102,057;
July 25, 1978; filed March 5, 1976.
45. K. Henning and J. Kandler (assigned to Hoechst AG), DE-OS 2,532,804;
February 10, 1977; filed July 27, 1975.
46. W. P. Fethke, Jr. (assigned to Procter & Gamble Co.), FR 1,557,535; Feb-
ruary 14, filed March 28, 1968.
47. H. Poselt, A. D. Tomlinson, and H. Rabitsch (assigned to Unilever N.V.),
DE-OS 3,144,470; May 10, 1983; filed November 9, 1981.
48. H. D. Hathaway and B. J. Heile (assigned to Procter & Gamble Co.), US
3,399,144; August 27, 1968; filed January 4, 1966.
49. M. A. N. Carter (assigned to Unilever N.V.), DE-OS 2,701,664; July 28,
1977; filed January 17, 1977.
50. D. Nickolls and W. B. Temple (assigned to Unilever PLC), EP 21,830; June
15, 1983; filed June 25, 1980.
51. M. Curtis, P.R. Garrett, and J. Mead (assigned to Unilever N.V.), EP 45,208;
October 24, 1984; filed July 27, 1981.
52. G. Butler, M. N. A. Carter, M. Curtis, and R. M. Davies (assigned to Un-
ilever PLC), EP 54,436; June 23, 1982; filed December 15, 1981.
53. R. L. Mayhew and F. Krupin, Soap Chem. Spee. 38(4): 55 (1962).
54. ibid.' p. 80.
55. H. Ferch, Chem. Ing. Tech. 48:11 (1976).
56. Ullmanns Encyclopiidie der technischen Chemie, Bd. 21, VCH, Weinheim,
1982, pp. 462ff.
57. K. Winnacker and L. Kiichler, Chemische Technologie, Bd. 3, Anorganische
Technologie II, 4th ed., Hanser, Miinchen Wien, I 983, pp. 75ff.
58. H. Kloepfer (assigned to Degussa AG), DE 762,723; February 28, 1942.
59. E. Wagner and H. Brunner, Angew. Chem. 72: 744 (1960).
60. L. J. White, J. G. Duffy, and G. L. Cabot, Ind. Eng. Chem. 51: 232 (1959).
61. K. Andrich (assigned to Degussa AG) DE 966,985; September 12, 1957;
filed September 13, 1951.
62. A. Becker and P. Nauroth (assigned to Degussa AG), DE 1,467,019; July
27, 1970; filed March 2, 1962.
63. E. M. Allen (assigned to Columbia-Southern Chemical Corp.), FR 1,134,178;
September 2, 1955; filed November 5, 1954.
266 Ferch and Leonhardt

64. R. Hoesch and W. A. Albrecht (assigned to Chemische Fabrik Hoesch KG),


DE I, 117 ,552; May 30, 1962; filed April 17, 1959.
65. B. Brandt, P. Nauroth, A. Peters, and H. Reinhardt (assigned to Degussa
AG), DE-OS 2,447,613; April 8, 1976; filed October 5, 1974.
66. B. Brandt, P. Nauroth, A. Peters and H. Reinhardt (assigned to Degussa
AG), DE-OS 2,505,191; August 26, 1976; filed February 7, 1975.
67. H. Brunner and D. Schutte, Chem. Ing. Tech. 89: 437 (1965).
68. H. Reinhardt, K. Trebinger, and G. Kallrath (assigned to Degussa AG), DE-
OS 2,435,860; February 12, 1976; filed July 25, 1974.
69. Roderburg (assigned to Chem. Fabrik Wesseling AG), DE-AS I, 074, 559;
February 4, 1960; filed June 25, 1959.
70. H. Winkeler, Kautsch. Gummi Kunstst. 30:6(1977).
71. L 0 Young (assigned to W. R. Grace & Co.), DE 1,006, 100; October JO,
1957; filed May 27, 1955.
72. 0. Kiihnert, G. Tiirk, and E. Eisenmenger (assigned to Degussa AG), DE
1,592,865; April 24, 1980; filed October 12, 1967.
73. A. H. Warth, The Chemistry and Technologie of Waxes, 2nd ed., Reinhold,
reprint, 1960, pp. 391-393, 421 ff.
74. P. Peltre and A. Lafleur (assigned to Procter & Gamble Co.), DE-OS 2,857, 155;
January 3, 1980; filed June 12, 1978.
75. R. E. Atkinson and D. A. Ross (assigned to Procter & Gamble Co.), EP
8,829; March 19, 1980; filed August 28, 1979.
76. H.-U. Hempel and E. Schmadd (assigned to Henkel KGaA), DE-OS 3, 115,644;
November 4, 1982; filed April 18, 1981.
77. L H. Tan Tai (assigned to Unilever PLC), EP 109,247; May 23, !984; filed
November 8, 1983.
78. J.C. Wuhrmann, W. Seiter, B. Giesen, and E. Schmadel (assigned to Henkel
KGaA), DE-OS 3,400,008; July II, 1985; filed January 2, 1984.
79. J.-P. Briand and C. C. Storer (assigned to Unilever N. V.), FR 2,559, 400;
August 16, 1985; filed February 12, 1985.
80. V. B. John, G. C. Sawicki, R. Pope, and R. J. Seampton (assigned to Dow
Coming Ltd.), EP 217, 501; April 8, 1987; filed July 23, 1986.
81. D. N. Willing (assigned to Dow Coming Corp.), EP 163,398; July 20, 1988;
filed April 15, 1985.
82. K. W. Farminer and Ch. M. Brooke (assigned to DOW Coming Ltd.), GB
l,378,874; December 27, 1974; filed July l, 1971.
83. G. C. Sawicki J. Am. Oil Chem. Soc. 65(5): 1013 (1988).
84. H. W. Fox, P. W. Taylor, and W. A. Pisman, Ind. Eng. Chem. 39: 1401
(1947).
85. D. Myers, Surfactant Science and Technologie, VCH, Weinheim, 1988, pp.
201, 204.
86. G. Bartolotta, N. T. De Oude, and A. A. Gunkel (assigned to Procter &
Gamble Co.), GB 1,407, 997; October l, 1975; filed August 1, 1972.
87. A. C. McRitchie (assigned to Procter & Gamble Ltd.), GB 1,492,939; No-
vember 2, 1977, filed March 11, 1974.
Foam Control in Detergent Products 267

88. A. Boeck, P. Krings, and E. Smulders (assigned to Henkel KGaA), DE-OS


2,753,680; June 7, 1979; filed December 2, 1977.
89. K. Bachmann, D. Jung, and A. Boeck (assigned to Henkel KGaA), EP 13,028;
March 10, 1982 filed December 24, 1979.
90. H.-G. Smolka, H. Reuter, M. Berg, and G. Vogt (assigned to Henkel KGaA),
EP 22,998; January 28, 1981, filed July 12, 1980.
91. M. Berg, G. Vogt, H.-G. Smolka and H. Reuter (assigned to Henkel KGaA),
EP 36, 162; September 23, 1981 ; filed March 9, 1981.
92. L. H. Tan Tai (assigned to Unilever PLC), EP 40,091; September 14, 1983
filed May 11, 1981.
93. H. Reuter, H. Saran, and M. Witthaus (assigned to Henkel KGaA), EP 70,491;
January 30, 1985; filed July 12, 1982.
94. S. Dhanani, R. Mac Donald, J. S. Clunie, and M. C. Brooks (assigned to
Procter & Gamble Co.), EP 46,342; February 13, 1985; filed July 23, 1981.
95. L. H. Tan Tai (assigned to Unilever PLC), EP 71,481; February 9, 1983;
filed July 27, 1982.
96. P.A. Morgan (assigned to Procter & Gamble Co.), EP 91,802; October 19,
1983; filed April 8, 1983.
97. L. H. Tan Tai (assigned to Unilever PLC), EP 94,250; July 30, 1986; filed
May 10, 1983.
98. H.-F. Fink, H.-J. Patzke, F. Spieker, and F.-J. Tolle (assigned to Th. Gold-
schmidt AG), EP 97,867; April 11, 1985; filed June 14, 1983.
99. H. Reuter and W. Seiter (assigned to Henkel KGaA), DE-OS 3,436,194;
April 10, 1986 filed October 3, 1984.
100. M. S. Gowland, S. A. Johnson, and R. Pell (assigned to Procter & Gamble
Co.), EP 142,910; May 29, 1985; filed August 21, 1984.
101. D. W. York (assigned to Procter & Gamble Co.), EP 163,352; December 4,
1985; filed May 22, 1985.
102. R. M. Baginski, B. C. Dems, L. A. Ross, and R. H. Soule, Jr. (assigned
to Procter & Gamble Co.), US 4,652,392; March 24, 1987; filed June 30,
1985.
103. R. Foret and L. H. Tan Tai (assigned to Unilever PLC), EP 206,522; August
9, 1989; filed May 25, 1986.
104. W. J. Hey and J. W. H. Yorke (assigned to Unilever PLC), EP 213,953;
September 20, 1989, filed September 2, 1986.
105. P. Schulz, J. Waldmann, F. J. Carduck, M. Witthaus, and E. Schmadel
(assigned to Henkel KGaA), DE-OS 3,633,519; April 14, 1988 filed October
2, 1986.
106. P. M. Burrill (assigned to Dow Coming Ltd.), EP 210,721; April 2, 1987,
filed May 6, 1986.
107. P. M. Burrill (assigned to Dow Coming Ltd.), EP 210,731; April 2, 1987,
filed June 9, 1986.
108. P. W. Appel, F. Bartolotti, F. Delwel, A. D. Tomlinson, S. Willemse, and
F. Hornung (assigned to Unilever PLC), EP 256,833; February 24, 1988,
filed August 10, 1987.
268 Ferch and Leonhardt

109. P.R. Garrett, M. Hewitt, W. J. Iley, P. C. Knight, A. P. Pilidis, L. H. Tan


Tai, T. Taylor, and J. W. H. Yorke (assigned to Unilever PLC), EP 266,863;
May 11, 1988; filed August 10, 1987.
110. J. W. Revis, J. A. Sagel, and D. K. Ostendorf(assigned to Procter & Gamble
Co.), GB 2,204,825; November 23, 1988; filed April 29, 1988.
111. A. Asbeck, A. Meffert, G. Rombey, and K.-H. Schmid (assigned to Henkel
KGaA), EP 301,412; filed July 21, 1988.
112. P. J. Chevalier (assigned to Societe des Usines Chimiques Rhone-Poulenc),
US 3,113,930; December 10, 1963. filed April 1, 1960.
113. S. Nitzsche and E. Pirson (assigned to Wacker-Chemie GmbH), US 3,235,509;
February 15, 1966; filed October 3,1982.
114. R. E. Patterson, 1988 Nonwoven Conference, Tappi Press, Atlanta, 1988,
pp. 39-48.
115. N. Nguyen, Evidence that Polydimethylsiloxane is a Hard Acid, M.S. Dis-
sertation, Rensswelaer Polytechnic Institute, 1985.
116. K. W. Farminer and Ch. M. Brooke (assigned to Dow Corning Ltd.), GB
1,450,580; September 22, 1976; filed January 22, 1973.
117. K.-H. Millier, R. Tailfer and G. Tiirk (assigned to Degussa AG), DE-OS
3,001,573; July 23, 1981; filed January 17, 1980.
118. G. Giesselmann and K. Giinther (assigned to Degussa AG), EP 62,748; May
15, 1985; filed February 16, 1982.
119. C. Atkinson, W. Jley, P. C. Knight, P. J. Russel, T. Taylor, and D. P.
Hones (assigned to Unilever PLC), EP 221,776; May 13, 1987; filed October
30, 1986.
7
Antifoams for Paints

MAURICE R. PORTER Maurice R. Porter and Associates, Sully, South


Glamorgan, United Kingdom

I. Introduction and Historical Background 270


II. Paint Systems 272
A. Water-based emulsion paints for household use 272
B. Industrial paint systems 274
III. Antifoams for Water-Based Household Emulsion Paints 276
A. Hydrocarbon/fatty acid/ester blends 276
B. Hydrophobic particle/hydrophobic oil mixtures 278
C. Water-based antifoam mixtures 281
D. Single component antifoams 283
IV. Antifoams for Industrial Paints 284
v. Adverse Effects of Antifoams 289
A. Wetting-out problems 289
B. Loss of color 290
C. Loss of gloss 291
VI. Tests for Antifoam Effectiveness 292
A. Paint manufacture 292
B. Can filling 293
C. Application 293
D. Stability on storage 294
VII. Speculations on Future Development 294
A. Water-based emulsion paints for household use 294
B. Industrial paint systems 295
References 295

269
270 Porter

I. INTRODUCTION AND HISTORICAL BACKGROUND


Paint consists of solid particles and a binder dispersed in water or solvent.
The particles give opacity and color. The binder binds the particles together
into a coherent paint film when either solvent or water has evaporated. The
main products of the industry are water-based emulsion paints for household
use, solvent-based household paints, and mixed water/solvent-based paints
for industrial use. The latter are used in applications such as automobile
finishing and refrigerator coatings. There is a slow but steady shift from
solvent-based paints to water-based paints both in Europe and the United
States. Foam is an especially severe problem in both water-based and water/
solvent-based paints particularly where emulsion polymers are used as bind-
ers. This account is in the main confined to the use of antifoams for water-
or water/solvent-based paints.
Foam problems in paints occur in manufacture, canning, and application.
During manufacture for example agitation often leads to air entrainment and
loss of capacity in manufacturing plants. Thus the manufacture of paints has
the following basic steps:
1. Grinding the solid particles to an appropriate size. This operation is
usually carried out at high shear rate.
2. Mixing in the binder usually carried out at medium to low shear rate.
3. Mixing in the pigment and other additives, usually carried out at low
shear rate.
The high shear rates required for many of these operations necessitates use
of, for example, high-speed impeller mills, sand mills, ball mills, or roller
mills. Although this equipment is usually designed to avoid foam formation
some air entrainment is an inevitable consequence of the intense agitation
involved.
Foam problems can also occur when filling paint cans. Here weight not
volume is the usual measure of quantity. Weight specifications cannot be
met if the paint contains excessive amounts of air.
Bubbles of air may be entrained during the application of paint which
can interfere with the quality of the coating. If, however, these bubbles burst
well before the paint dries, then the paint film will be unaffected. Defects
in the film due to air bubbles are therefore associated with the presence of
bubbles at or near the point of solidification of the coating. If bubbles burst
at the point of solidification, then either small circular defects known as
"pin holes" due to small bubbles or large irregular defects due to collections
of polyhedral bubbles known as "craters" are found. If bubbles do not burst
at all, then a reduction in the reflectivity of the coating (i.e., "gloss") may
occur.
Antifoams for Paints 271

In water-based emulsion paints stabilization of foam derives mainly from


the surfactants and polymeric stabilizers used in the manufacture of the paint
(see Gress [l ]). Thus high-foaming anionic surfactants are used in the prep-
aration of emulsion polymers from which synthetic water-based emulsion
paints are made. Industrial paints, on the other hand, often employ water-
cosolvent mixtures so that the polymeric binder is rendered soluble. This
combination of ingredients can yield relatively stable foam inespective of
the presence of surfactant contaminants [2].
The most generally applicable approach to foam problems is utilization
of the appropriate antifoam technology. For water-based emulsion paints
where foam problems derive from surfactant contaminants there exists an
alternative approach based upon utilization of low-foam surfactants. How-
ever, such materials are usually unacceptably expensive.
In the early days of antifoam development for paints it was expected that
one type would be developed which would be acceptable for all water-con-
taining paints. Antifoams consisting of mixtures of hydrophobed silica and
polydimethylsiloxanes (silicone antifoams) were once considered to satisfy
this criterion. However, coating defects induced by the presence of poly-
dimethylsiloxane droplets were so severe that silicone antifoams were at one
time eliminated from water-based paints altogether.
The next phase in the development of antifoam technology for paints con-
cerned discovery of the specificity of antifoams. Thus it proved extremely
difficult to make an anti foam which was efficient over a wide range of paint
formulations. At the same time similar problems were occurring in the tex-
tile industry where the same high-foam synthetic surfactants were used. Here
foam problems were occurring particularly in the scouring of cotton and
wool and the disperse dyeing of polyester fibers. The paper industry was
also suffering from considerable foaming problems due to the introduction
of new synthetic sizes. Both the paper industry and textile industry needs
were far larger than the paint industry. Companies connected with paper and
textiles developed a large number of antifoams specifically for textiles and
paper. These products were then tried in paints and often worked. Thus the
development of antifoams for paint, textiles, and paper developed along sim-
ilar but not identical lines. Companies specializing in antifoams would offer
many new products every year mainly based on blends of natural oils, fatty
acids, esters, petroleum oils and waxes. Early products were difficult to both
handle and disperse despite their cost effectiveness. Over the years anti-
foams, which in some measure overcome these handling and dispersion dif-
ficulties, have been developed. However, there remains considerable scope
for further improvement and much development work is still directed at these
issues.
272 Porter

The latest phase of antifoam development concerns the use of block co-
polymers of siloxanes and, for example, polyoxyethylene and polyoxypro-
pylene compounds. These antifoams are sometimes used alone so that no
mixing or blending is required. They are therefore easier to manufacture
reproducibly.
All these different aspects of the development of antifoams for paints are
still proceeding, but the overall driving force is cost effectiveness. The re-
duction of foam does not usually add any positive properties to the product,
and the paint manufacturer will be looking for the most economical way to
eliminate foam from his products.

II. PAINT SYSTEMS


A. Water-Based Emulsion Paints for Household Use
A typical water based emulsion paint [3] is given below. The constituents
listed are mixed in the sequence shown.
Parts by
Weight
Propylene glycol 6.52
Aqueous polyacrylate solution (25%) 1.40
Antifoaming agent 0.47
Water 1.86
Titanium dioxide pigment 26.53
Water 4.66
Aqueous polyacrylate latex dispersions (46%) 50.90
Tributyl phosphate 1.71
Preservative 0.28
Antifoaming agent 0.47
Ammonia (25%) 0.20
1,2-Prophyleneglycol 2.80
Ammonium polyacrylate solution (25%) 0.93
Water 2.20
100.00
In special cases an additional antifoam may be added under conditions of
low shear just before filling the cans. Some antifoams in particular paint
systems when added at this late stage are extremely efficient at reducing
foam produced by a brush or roller. The same antifoam added at an earlier
stage can be quite ineffective.
Surfactant may be added to help wetting in of the solid pigment. Sur-
factant will also be present in the emulsion polymer (the "latex"). It is these
Antlfoams for Paints 273

surfactants which are mainly responsible for stabilizing foam. The thickener
(polyacrylate or hydroxyethyl cellulose) will also help to stabilize any foam
that is formed. Thus the more viscous the paint the more stable the foam
even in the presence of antifoam (see Chapter 1). The dispersing agent will
be a low molecular weight sodium polyacrylate or polyphosphate. These
materials do not themselves stabilize foam, but enhance the tendency of
certain surfactants to do so.
The main surfactants which cause foam are those present in the emulsion
polymer. Emulsion polymers are dispersions of discrete polymer particles
(0.04-3.0 micron) in water. They are prepared from monomers, water, sur-
factants, and initiator. Other common, but not necessarily essential ingre-
dients, are water-soluble polymers which act as protective colloids. An ex-
cellent account of the role of surfactants in emulsion polymerization is given
by Groves [4].
Surfactants used by emulsion polymer manufacturers are usually selected
to facilitate the emulsion polymerization process, to achieve a desired emul-
sion particle size and distribution, and to enhance storage stability of the
emulsion. Avoidance of foam in the finished paint will have low priority.
The nature of the paint film is determined by both the physical nature of
the polymer and the size and size distribution of the particles. The choice
of monomers (for example vinyl acetate, ethyl acrylate, 2-ethyl-hexyl ac-
rylate, butyl acrylate, styrene, ethylene, butadiene) will determine the rheo-
logical and adhesive properties of the coating. The size and distribution of
the particles will also determine the rheology of the paint and can affect
other properties such as rate of drying, rate of absorption by the substrate,
and the amount of filler which can be incorporated.
The type of surfactants normally used in emulsion polymerization are
those anionic and nonionic surfactants which are typically found in detergent
products. Anionic surfactants have found the most widespread use in emul-
sion polymerization.
Anionic surfactants yield electrical double-layer stabilization of the emul-
sion. Such surfactants will therefore show sensitivity to ionic strength as
determined by the presence of other salts in the system. Some typical anionic
surfactants used for emulsion polymerization include sodium lauryl sulfate,
sodium dodecyl benzene sulfonate, sodium a-olefin sulfonate. These yield
fine particle size emulsions (0.04-0.1 micron (4)) and high foam. Other
anionic surfactants such as sodium 2-ethylhexyl sulfate and sodium lauryl
ether (3EO) sulfate yield large particle size emulsions (0.1-3 microns[4])
and lower foam. Anionic surfactants are typically present at levels of 1-3%
as a proportion of the emulsion polymer.
Nonionic surfactants stabilize emulsion particles by surrounding the par-
ticles with a hydrated headgroup layer. This layer acts as a steric barrier to
274 Porter

particle coalescence. Nonionics are generally less efficient in stabilizing the


emulsion than anionics. They are however necessary to provide stability against
electrolytes and freeze-thaw cycles. Nonyl phenol ethoxylates (with 30%
ethylene oxide) and lauryl alcohol ethoxylates are typical examples both of
which are low foaming and yield large particle size emulsions. Typical con-
centrations would be 0.4-1.0% expressed as a proportion of the emulsion
polymer.
In practice a combination of anionic and nonionic surfactants is used to
give optimum properties. The paint manufacturer will use small particle size
emulsion polymers for most paints. As we have seen, surfactants which
yield small particle sizes also produce high foam.

B. Industrial Paint Systems


A typical household emulsion paint can be readily described but a typical
industrial paint cannot because industrial paints vary in composition to a far
wider degree than household paints. The essential features, however, remain
the same, namely a pigment, a binder, and solvent or mixed water-solvent
medium which is usually volatile. The binder can be an emulsion polymer
as described above. More usually, however, it is a solvent soluble polymer.
The paint film can be formed from a thermoplastic polymer, an air-drying
polymer or a heat cured thermosetting polymer. Industrial coatings known
as varnishes are clear transparent films which contain no pigment or very
low levels of pigment.
Thermosetting water-based paint systems tend to give the most severe
foam problems during application particularly if they are cured by heat or
catalyzed to cross-link very quickly. Examples of such systems are based
on resin binders such as

Water-reducible alkyds*
Epoxy resins
Acrylic colloidal emulsions
Polyesters

Each of these resin binders can be cross-linked with an amino-formaldehyde


compound using a reactive proton on the resin and a methylol (--CH 20H)
group on the amino-formaldehyde compound. The system can be catalyzed
with an organic acid (para-toluene sulphonic acid) so that the reaction may
be accelerated and a coating can be dried in seconds. If foam occurs during
application, it will have very little time to collapse so the essential function

*Alkyds are cross-linked polyesters produced from natural oils reacted with di basic acids and
polyols.
Antifoams for Paints 275

of any antifoam is to prevent foam from forming rather than ensuring any
foam formed collapses quickly.
A typical composition of a thermosetting water-based paint is given by
Nielsen [5]. Thus a water-reducible alkyd resin is mixed at low speed with
other ingredients to give a composition.

Parts by
weight
Medium oil length alkyd resin 100.5
Water 201.0
Ammonium hydroxide 8.0
Ethylene glycol butyl ether 11.0
Manganese catalyst 2.1
Titanium dioxide 287.7
Antifoam 10.0

The important feature of this formulation is a mixed solvent, ethylene-glycol


butyl ether/water, which is necessary to keep the alkyd resin in solution.
Depending upon the application method and the alkyd resin a wide variety
of solvent compositions are used in practice.
Water-based industrial paints using cosolvents differ from water-based
household paints in that the polymeric binder is dissolved in the former but
not in the latter. With such industrial paints Kushnir et al. [2] have shown
that the resin itself is the agent responsible for foaming. Thus at a certain
ratio of water to ethylene glycol butyl ether cosolvent a maximum in foam
stability is found. This correlates with both a maximum in viscosity and the
point of incipient phase separation of the polymeric resin.
Foam control in industrial paints is in general more difficult than with
household paints. This arises from a variety of causes. One of which con-
cerns rapid drying films. Other difficulties derive from more severe criteria
concerning the quality of the paint film where a smooth continuous film is
often essential. Defects are therefore more noticeable. Moreover, the per-
formance required of the paint with respect to for example corrosion resis-
tance is more critical. Therefore paint film defects due to either foam or
even antifoam are much less acceptable than with household paints.
Industrial paints are generally applied using more vigorous methods than
are household paints. Thus industrial paints are applied by for example
spraying, flood-coating, electrocoating, dipping, curtain coating, and ma-
chine reverse roller coating. Paints to be sprayed must be carefully formu-
lated to avoid foam. Machine roller coating can give problems due to foam
slowly being generated in the machine and then being suddenly transferred
276 Porter

to the reverse roller. The result is a coating which is subject to disastrous


deterioration in quality.
In addition to all these factors other difficulties for the antifoam formu-
lator arise because of the greater variety of industrial paint types which are
often tailored to specific applications with a wide variety of substrates (ma-
rine coating or car finishing, for example).

Ill. ANTIFOAMS FOR WATER-BASED HOUSEHOLD


EMULSION PAINTS
Development of antifoams for aqueous emulsion paints has followed essen-
tially empirical lines. As we have seen, there has been a need for formu-
lation of specific antifoams for specific paints. Often these formulations are
complex and contain many components. However a number of features are
common to most. Thus the antifoam ingredient must be insoluble and well
dispersed in the medium for which foam control is desired. In the case of
aqueous media the active antifoam ingredient must also be hydrophobic (see
Chapter 1). Typically hydrocarbons, siloxanes and fluorocarbons form the
basis of these hydrophobic ingredients. Often, however, these ingredients
are difficult to disperse and they are therefore sometimes predispersed to
form emulsions in water. Such emulsions are readily mixed with the foaming
medium to produce a fine dispersion of the active hydrophobic antifoam.
Surfactants are used to assist preparation and stabilization of these emul-
sions. Dilution of the antifoam in this manner, however, increases handling
costs and can produce storage problems. Dispersal difficulties are therefore
often overcome by adding a suitable surfactant to the antifoam ingredient
so that it may be more readily directly dispersed in the foaming medium.
Surfactants used to prepare emulsions or assist direct dispersal of antifoam
can, however, contribute to the foaming problem which the antifoam is sup-
posed to combat.
A. Hydrocarbon/Fatty Acid/Ester Blends
Blends consisting of hydrocarbons with fatty acids and various other ingre-
dients formed the basis of the early antifoams used in the paint industry.
These materials were formulated more or less without regard to any scien-
tific principles. They were often prepared as oil-in-water emulsions. The
earliest antifoams were usually pastes or soft solids which were difficult to
disperse directly for a given application often requiring a predispersal step
before addition to a system. Later products tended to be essentially liquid.
The rheological properties of these antifoams were largely dominated by the
physical state of the hydrocarbon. Thus liquids contain liquid hydrocarbons
and the pastes contain hydrocarbon waxes.
Antifoams for Paints 277

A typical example of an early paste was described by McGinn [6] over


40 years ago. This was based upon an empirical finding that a mixture of
a mineral wax and a partial glyceride of a fatty acid was a good antifoam
in paper manufacture but that addition of a ricinoleic acid gave improved
performance. The composition for paints derived by McGinn [6] from this
consisted of
Parts by weight
Scale wax 20
Candelilla wax 2
Glyceryl monostearate 15
Ricinoleic acid 5
This was mixed until uniform and then the following added at 50°C and
stirred
Parts by weight
23.5% sodium hydroxide 2.3
Water 55.5
Finally 0.2 part by weight of formaldehyde was added as preservative.
The waxes are readily recognized as the main hydrophobic components
of this antifoam blend. The ricinoleate soap, formed in situ, is a dispersal
aid, and the glyceryl monostearate probably functions as a hydrophobic
component of the antifoam. This early example illustrates the complex na-
ture of antifoam blends used in emulsion paints. A similar patent by McGinn[7]
teaches the use of glyceryl monostearate and methyl oleate.
Ten years later Snook [8] modified the same basic composition to en-
hance fluidity. This was done by first mixing the following components at
60°C.
Parts by weight
Scale wax 5
Paraffin oil 5
Glyceryl monostearate 5
Methyl ricinoleate 10
Monostearate 10
To this blend was added
Parts by weight
Water 64
40% aq. formaldehyde 0.4
Later developments produced clear liquid products from which water was
excluded. An example, due to Schott and Ward[9], consists of
278 Porter

Paraffin oil 61.5


PPGEO 9.75
Peanut oil/diethylene glycol reaction product 9.75
Polyethylene glycol (200 mol wt.) coconut fatty acid 11.6
Tall oil 5.85
Triethanolamine 1.55
Here PPGEO is a butyl-blocked polypropylene glycol ester of oleic acid.
This concoction is typical of antifoams formulated in the sixties. Many
similar concoctions are used today because of advantages of cost effective-
ness and storage stability. Modem products often contain polysiloxane poly-
glycol copolymers which can boost performance as described by Gruenert
et al. [10].
With such complex concoctions it is doubtful whether even present pro-
ducers have a clear understanding of the function of each component and
even whether it is really necessary. These concoctions are however very easy
to manufacture, capable of an enormous number of variations and are ex-
tremely difficult to analyze.

B. Hydrophobic Particle/Hydrophobic Oil Mixtures


That finely divided hydrophobic particles may significantly enhance the an-
tifoam performance of hydrophobic oils represents an important discovery.
Silica had been incorporated into silicone antifoams as far back as 1953 [ 11]
but the importance had not been appreciated. Boylan in the early sixties [12]
found that hydrophobic silica increased the antifoaming efficiency of both
polydimethylsiloxane and organic hydrophobic liquids in a dramatic manner.
A typical example is given by Liebling and Canaris [13] where mixtures of
polydimethylsiloxane (1-6 parts by weight) and silica (8.9 parts by weight)
were mixed to a dry powder and heated to 196°C to form hydrophobed silica.
An effective antifoam was prepared by adding this to mineral oil (89.5 parts
by weight).
An essential feature of this type of antifoam is that the silica surface be
rendered hydrophobic. There are many ways of achieving this. The method
described by Liebling and Canaris [13] needs prolonged heating to high tem-
peratures. Other methods using different reactive species can involve less
severe conditions. Examples of such species include
Polysiloxane fluid with Si-OH end groups (see Sullivan [14])
Organo silicon halides[15]
Cyclic polysiloxanes[ 16]
Trimethylsilylamines[ 17]
Hexamethyldisilazane[ 18]
Antifoams for Paints 279

Other finely divided solids will also function in this context. Examples
include

Polyvinyl chloride [ 19)


Polypropylene [201
Etylenebis( stearamide) [21]
Amide treated with chlorsilanes [22)
Talc [23]
Long-chain aliphatic ketones and secondary alcohols [24]
Hydroxy stearyl alcohol fatty acid esters [25]

Small particle size and a strongly hydrophobic surface are essential require-
ments for these materials if they are to be effective ingredients in admixture
with hydrophobic oils for aqueous systems. The most effective and versatile
particles have so far proved to be silica hydrophobed with polydimethylsi-
loxane. However, other types of particles have often found application for
paints. Thus for example the ester of hydroxystearyl alcohol with saturated
fatty acid (of chain length C 18 -C22 ) finely dispersed in mineral oil [25)
forms a particularly effective antifoam for emulsion paints. It seems possible
that many of the early antifoam compositions based on hydrocarbon/fatty
acid/ester blends (Sec. III.A.) are in fact variations of the simple concept
of hydrophobic particle/hydrophobic oil mixtures modified by the addition
of dispersing agents and other ingredients of sometimes dubious usefulness.
A detailed description of the properties of mixtures of hydrophobic particles
and oils as antifoams together with hypotheses concerning their mode of
action is given in Chapter 1.
Mixtures of different finely divided hydrophobic solids are also described
in the patent literature. An example is talc and ethylenebis(stearamide) mixed
with a hydrocarbon oil. Thus Curtis and Woodward[23] teach a formulation
consisting of

Parts by weight
Liquid hydrocarbon 74.8
Vinyl acetate/fumerate ester of tallow 7
alcohol
Talc 15
Etylenebis( stearamide) 2
Silicone oil 0.2

All the components are heated together at 160°C, cooled and homogenized.
This patent claims that the bis(stearamide) (which is insoluble) helps to slow
down the settling of the talc and enhance the activity of the antifoam. Par-
280 Porter

ticle settling is, however, still a major problem with this type of defoamer.
Vinylacetate/fumarate copolymer acts as both dispersing agent for the par-
ticles and as thickening agent for the liquid hydrocarbon (which inhibits
settling).
Use of hydrophobic silica with other finely divided solids also apparently
produces efficient products. Sinka et al. [26] compare a combination of hy-
drophobic silica/ethylenebis(stearamide) in mineral oil to that of hydropho-
bic silica alone in mineral oil in latex paints when mixed on a paint shaker.
Results are presented below.
Product % entrained air
Blank (no antifoam) 17.00
Hydrophobic silica 1.37
Hydrophobic silica/ 0.55
ethylenebis(stearamide)
Although antifoams based upon hydrophobic silica/ethylenebis(stearamide)
were found to be more efficient it was found that the hydrophobic silica
would begin to settle out and the viscosity would change with time. Im-
proved stability was obtained by addition of polyoxypropylene glycols with
butyl end groups and organophosphate esters. However, the most stable
products were generally inferior in antifoam efficiency to less stable products.
Apart from enhanced antifoam efficiency another advantage associated
with addition of hydrophobic silica to ethylenebis(stearamide) concerns the
rheological behavior of the antifoam. Thus antifoams with high concentra-
tions of ethylenebis(stearamide) in mineral oil are very efficient but are gen-
erally of very high viscosity and are therefore difficult to handle and pump.
It has been claimed by Michalski and Youngs [27] that hydrophobic silica
can act as a viscosity reducing agent in such systems.
It is clear then that a major problem with hydrocarbon oil antifoams con-
taining finely divided solids is the stability of the composition. During stor-
age solids can settle out. This process is sometimes accentuated by decrease
in viscosity. Ihde [28] describes the shelf stability problems of silica-based
antifoams and the addition of oil soluble organic resins such as coumarone-
indene, petroleum hydrocarbon, terpene resin, or hydrocarbon-formaldehyde
resins to stabilise and inhibit settling of the silica.
There have been many patents filed on the use of the thickening and
gelling agents. However, when complex multiphase systems are manufac-
tured then the method of mixing, rates of heating, rates of cooling and the
type of equipment used all become relevant in determining the efficacy of
the antifoam. The following example by Lichtman and Rosengar [29] illus-
trates the complexity of some products. Their preparation commences with
a mixture of
Antifoams for Paints 281

Parts by weight
Paraffin oil 6
Castor oil 1
Ethylenebis( stearamide) 4
which is heated at 140°C. To this mixture is then added vinyl acetate/fu-
marate (3 .5 parts by weight) as gelling agent. This concoction is maintained
at 140°C for 15 min and then added rapidly with agitation to paraffin oil
(85.3 parts) at l5°C. The resulting mixture is cooled to 26°C and subse-
quently homogenized at a certain pressure. The temperature is then raised
to 50°C for 4-12 hr with no agitation followed by cooling to 25°C with
agitation. Silicone oil (0.2 part) is then added. It is claimed [28] that this
composition yields a good antifoam performance in latex paints.
Despite difficulties concerning stability antifoam systems based upon
mixtures of hydrophobic particles and hydrocarbon oils are extremely effi-
cient and relatively cheap. Many combinations of the basic ingredients are
possible. Patents are still being filed with novel combinations (see, for ex-
ample, Kaychok and Boylan [30], Flannigan[31], and Wuhrmann et al. [32]).

C. Water-Based Antifoam Mixtures


Water-based antifoam systems are usually oil-in-water emulsions where the
main advantage over use of the neat antifoam concerns ready dispersion in
the medium to be defoamed. A difficulty with this approach concerns prep-
aration of a stable emulsion which when diluted becomes unstable in order
to drive the active antifoam ingredient (the oil) to the air-water surface and
not remain in the bulk of the system. Mixtures of hydrophobic particles and
hydrophobic oils do in fact show a tendency to form water-in-oil emulsions
(see Chapter 1). Water-based antifoam systems claimed to be suitable for
latex paints are also prepared in the form of such inverted emulsions[33].
Here water-in-oil emulsions presumably disperse less readily in water than
the neat antifoam and incur higher storage and transport costs per unit weight
of active ingredient. Stability to sedimentation on storage may, however, be
greater than for the neat antifoam.
An example of an early oil-in-water antifoam is [34]
Parts by weight
C 16 fatty alcohol 35
Hexadecane 5
White paraffin oil 3
C11H3sCOO(E0) 10H 2
Water 65
Using suitable emulsifying techniques the mixture gives an oil droplet size
282 Porter

distribution which is optimal for both shelf stability and antifoam efficiency.
This system has however proved more successful in papermaking applica-
tions than in paints but represents an important development showing the
need for the correct droplet size distribution.
Later oil-in-water emulsions incorporated hydrophobed silica. Details of
a method for preparation of water-based antifoams from silicone oil and
hydrophobed silica are, for example, given by Raleigh [34].
Water-based products have generally simply involved emulsification of
oil-based antifoams which are known to perform efficiently. Although the
approach has been effective the resulting formulations have now become so
complicated that it is often difficult to elucidate the function of each com-
ponent. For example, some water-based formulations now contain

Active antifoam ingredients


Mineral oil
Hydrophobic silica
EO /PO copolymers
Silicone fluid
Dispersing agents
Polyethylene glycol monoleates and dioleates
Naphthalene sulfonates
Glyceryl stearates
Nonyl phenol ehthoxylates
Stabilizers (to give the emulsion storage stability)
Xanthan gum
Polyacrylic acid salts
Vinyl acetate/maleate copolymers
Preservative
Formaldehyde

An example of such a complex formulation is given by Schmidt and Gam-


mon [33). These inventors claim a water-in-oil antifoam blend suitable for
latex paints. This is prepared by first mixing

Parts by weight
Paraffin oil 2
Ethylenebis( stearamide) 2.5
Polymethacry late 3
Silicone surfactant 1.5

and heating to 150°C. The hot mixture is then added to paraffin oil (21 parts
by weight) at 25°C and the whole cooled to 50°C. This is then added to
Antifoams for Paints 283

Parts by weight
Paraffin/naphthenic oil 5
Paraffinic oil 6
Hydrophobic silica in oil 8
Castor oil + 15EO
Water 49.0
Formaldehyde 0.1
The resulting mixture is homogenized and silicone oil (0.1 part) added.
Careful specification of the order of addition of components, rates of heating
and cooling at various stages, and shear rates during mixing all suggest that
these factors affect the performance of the resulting antifoam. Rapid cool-
ing, for example, can have a significant effect on the particle size of any
component which precipitates out of solution. This will in tum affect the
antifoam efficiency and storage stability (settling rates) if, for example, that
component is the hydrophobic particle component of the antifoam.
Sedimentation on storage is still a potential problem with these multiphase
antifoam concoctions. In an attempt to eliminate this problem, Gammon [361
prepared oil-in-water microemulsion antifoam concoctions. These are, of
course, optically clear liquids which do not exhibit phase separation and
therefore present no potential storage problems. They consist of a water
insoluble hydrophobic organic antifoam material (ranging from triglycerides
to hydrocarbons) mixed with an ethoxylated nonionic surfactant, a "cou-
pling" agent (such as an alkylene glycol), water, a base, and an antigelling
agent. Gammon [36] claims that such concoctions are effective antifoams
for paints. However, few microcmulsion products have appeared on the mar-
ket to date.

D. Single Component Antifoams


Potential storage problems are, as we have seen, always possible with mul-
tiphase antifoam systems. Use of a single-phase system avoids many of these
problems.
The microemulsions described by Gammon [36] are clearly representative
of such an approach. Other examples involve the use of single component
antifoams. In the main this means polymeric materials containing a signif-
icant proportion of polyoxyethylene or polyoxypropylene or both as block
copolymers. When dispersed in solution at temperatures above their cloud
points, drops of cloud phase separate out which function as antifoams (see
Chapter 1).
Simple polyoxyethylene/polyoxypropylene block copolymers and their
derivatives with a hydrocarbon chain have found application for many years
in industrial fermentation, industrial detergents, and machine dishwashing
284 Porter

(see Chapter 9). They have, however, not proved very successful as paint
antifoams. This may well concern the effect of the significant concentrations
of anionic surfactant, commonly present in emulsion paints, on the cloud
points of these materials. Thus it is well known that the cloud points of
ethoxylated materials increase with the addition of anionic surfactant [37].
If the cloud point is increased above the application temperature, then any
foam control will be lost.
These problems can be largely overcome with copolymers of polyoxy-
alkylenes and polysiloxanes. Such polyalkoxypolysiloxane copolymers will
function as antifoams for emulsion paints but can give paint film defects
similar to those found with polydimethylsiloxanes (see Sec. V). Steinberger
and Werner [38] claim to have produced polyalkyoxypolysiloxane copoly-
mers which do not give such defects. However, these materials require the
presence of hydrophobed silica in order to enhance antifoam performance
in much the same manner as is found with hydrocarbons or polydimethyl-
siloxanes. They therefore cannot be regarded as single component antifoams.
Fink et al. [39] describe another polyalkoxypolysiloxane which can be
used alone. These materials, however, only provide deaeration and do not
effectively destroy foam in emulsion paints.
Polyalkoxypolysiloxanes have not so far replaced the older mineral oil
based antifoams such as for example the mineral oil/wax/emulsifier/hy-
drophobic silica combinations for emulsion paints. They are both more spe-
cific in application and more expensive. These materials are, however, prov-
ing more successful in industrial paints (see Sec. IV).

IV. ANTIFOAMS FOR INDUSTRIAL PAINTS


Here we remember that the industrial paints for which foam problems are
most severe are those consisting of mixtures of water and a cosolvent. These
water-based industrial paints are distinguished from household emulsion paints
in that the binder is usually dissolved in the water-cosolvent mixture. Vari-
ations in solvent composition will then affect the viscosity and surface prop-
erties of the solution and therefore the foam properties and resistance to
antifoam. Kushnir et al. [2] have, for example, examined the foam behavior
of a water-based coating prepared with an acid functional acrylic polymer
(typical of those used in flow coat/dip coat formulations) cross-linked with
melamine. Most of this work was done with varnishes (i.e., paint without
pigment) using a typical formulation of 18%-20% resin, 7%-12% cosolvent
(ethylene glycol butyl ether) dimethyl-ethanol amine mixture and 65%-75%
water. Interestingly it was found that the cosolvent level had a marked effect
on foam behavior. Thus foam stability was very low at high cosolvent levels
(very soluble systems) then increased as the cosolvent level was reduced,
Antifoams for Paints 285

passed through a maximum and then decreased again with further reduction
of cosolvent. This foaming maximum correlated with a viscosity maximum
and an abrupt change in dynamic surface behavior as revealed by the phase
angles of dynamic dilational moduli. These authors attribute this behavior
to the onset of phase separation of the resin binder as the cosolvent level is
reduced and where large aggregates are forming in solution. This gives rise
to enhanced viscosity, slow foam film drainage rates and an enhanced re-
sistance to antifoam as revealed by tests with seven different antifoams.
Dramatic increases in the phase angles of dynamic dilational moduli will be
associated with the formation of such large aggregates [40]. Kushnir et al.
[2] state that defoaming a coating system near this foaming maximum is
impractical.
Other authors have emphasised the importance of the solubility of the
antifoam in the paint. Thus Schnall [41] carried out tests using unspecified
antifoams on an air-drying alkyd, a baking polyester-melamine, and a bak-
ing acrylic-melamine where each formulation had five different ratios of
water to organic solvent. In most cases it was shown that antifoam efficiency
varied with the solvent/water ratio so that the less soluble the antifoam the
higher the efficiency. Schnall [41] notes that if the antifoam dissolves it may
even enhance foam stability. Nielsen [42] also finds that changes in the
water/solvent ratio in a water-reducible alkyd could improve the antifoam
efficiency. The same author [5] compared eight different antifoams in a water-
reducible alkyd with different cosolvents and found that in many cases the
relative efficiency of the antifoams was dependent upon the cosolvent em-
ployed. It is not clear whether this is due to changes in antifoam solubility
or is due to changes in viscosity as found by Kushnir et al. [2].
The most successful antifoams for industrial paints are silicone based.
The most common and widely used silicones are the polydimethylsiloxane
fluids which generally, but not always, contain hydrophobic silica. In the
past these materials have given severe paint defects. There have however
been significant improvements in silicone-based products to avoid these de-
fects. These improvements have involved three different approaches.
The first utilizes a better knowledge of the relationship between the chem-
ical structure of the silicone and the observed paint defects. Much of this
information is confidential to the antifoam manufacturers. Dimethylpolysi-
loxanes, of molecular structure shown in Fig. 1, give the most efficient
defoaming but can also give the worst paint defects. They are available in
a wide range of molecular weights with n varying from l to several thou-
sand. Antifoaming efficiency becomes greater as n increases but so do paint
defects. By choosing the optimum value of n for particular paint formula-
tions, the paint defects can be reduced while retaining antifoam efficiency.
The optimum value of n will vary depending upon the paint system in gen-
286 Porter

CH3
I
CH3
I I
1 CH 3
I
C~ - Si - 0 - Si -10 - Si - CH
.3 I I I 3
CH3 c~] CH3
····n
FIG. 1 Polydimethylsiloxanes.

eral and the solvent system in particular. Although it is difficult to gener-


alize, the lower the molecular weight, the fewer the problems. However
there is a lower limit of molecular weight where the silicone does not act
as an antifoam, this is at about 40 cS. Careful formulation with other non-
silicone antifoams (e.g., fatty acid esters) can give excellent results.
A second approach utilizes dimethylpolysiloxane emulsions with con-
trolled particle size. Most paint defects attributable to silicones in water-
based systems are due to aggregation of the silicone droplets to form spots
which are the source of non wetting. Particles greater than 50 microns in size
will give problems. If the silicone emulsion has a very small particle size
which is stable and does not aggregate in use, then silicones are less prone
to give problems. However, if the particles are too small then antifoaming
efficiency falls away. Thus there must be a consistent controlled particle
size generally between 2 and 50 microns. This is achieved by using a com-
bination of the correct choice of surfactants and technique in emulsification.
Photomicrographs are shown in Figs. 2 and 3 showing a small particle (2-
30 micron) silicone emulsion compared to a coarse particle size (2-100 mi-
cron). Not only are the particles smaller in Fig. 2 but there is a smaller
spread of particle size [43].
The third approach to improvement of silicone antifoams for industrial
paints involves modification of the siloxane polymer to obtain better com-
patibility with the resin system. Thus silicone copolymers with polyesters,
amino-resins, urethanes, vinyls, and polyglycols show better compatibility
with resin systems which have a common chemical grouping. For example,
a siloxane copolymerized with a polyester is likely to be more compatible
with a polyester than the parent siloxane.
The most common type of modified siloxane are those block copolymers
formed from polydimethylsiloxanc and polyethylene oxide or polypropylene
oxide/ethylene oxide. As we have seen, these materials find occasional use
in water-based household paints. They are also used in industrial paint sys-
tems. Thus, for example, Berger et al. [44] describe their use as dcaerating
agents for low cosolvent and cosolvent-free hardenable resin paints.
Antifoams for Paints 287

FIG. 2 Photomicrograph of small-particle silicone emulsions (aqueous phase dyed


with methylene blue, magnification 90X).

FIG. 3 Photomicrograph of large-particle silicone emulsions (magnification 90 x ).


288 Porter

CH3
I
0 - Si - OSi( CH3)
I 3.
H
40

1
CH 3 CH 3
I I
(CH3 )3 Si 0-Si- 0-Si- 0Si(CH 3 ~

Jt,
I I
(CH 2) (CH2)10
I
co
I
CH 3

l
FIG. 4 Modified polysiloxanes.
(40-x)

Modified siloxanes may, in principle, be tailored to produce surface and


solubility characteristics which are compatible with any cosolvent/water-
based paint system. Thus antifoam behavior may be optimized and paint
film defects, due to preferential wetting of the substrate by the siloxane,
minimized. For industrial cosolvent/water-based coatings hydrophobed sil-
ica is not usually mixed with these modified siloxanes.
An example of the use of this approach for industrial paints is given by
Hempel et al. [45], who describe a typical method of introducing long-chain
alkyl groups, ester groups, and polyglycols into a polysiloxane chain. The
procedure is illustrated in Fig. 4. Here polydimethylhydrogensiloxane is re-
acted with a mixture of olefin and unsaturated ester using a platinum cata-
lyst. The ester group may then be transesterified with a monoalkylated poly-
ethylene glycol. A large number of variations are possible so that the solubility
Antifoams for Paints 289

and surface characteristics can be modified by varying either the olefin, ester
or polyglycol or their relative proportions. These materials are claimed by
Hempel et al. [45] to be effective antifoams for an air-drying water-reducible
alkyd, a polyacrylate, and a baking enamel.

V. ADVERSE EFFECTS OF ANTIFOAMS


A. Wetting-Out Problems
Antifoams added during manufacture of paint should ideally reduce foam
problems during manufacture, reduce aeration during can filling and reduce
foam problems during application. However, defects in the final dry paint
film can be caused by antifoams giving wetting-out problems in the wet coat
during application and subsequent drying. These defects can apparently look
very similar to defects in the paint film caused by air bubbles. Thus, defects
in the paint film can be caused either by the antifoam not working efficiently
so that bubbles are present or by the antifoam giving wetting-out problems.
There is considerable confusion in the literature concerning these defects.
Thus they are variously described as "cissing," "crawling," "fish-eyes,"
"craters," and "pinholes." These terms are not clearly defined. They are
generally ascribed to wetting-out problems, but this is not always true, as
small pinholes and larger "craters" can be due to air bubbles.
Water-based paints are replacing organic solvent-based coatings in both
household decorative paints and industrial coatings. Such water-based coat-
ings not only give significantly more foam but also less readily wet sub-
strates. The author has found that this latter problem increases in severity
in going from emulsion paints to semigloss emulsion paints to gloss emul-
sion paints and then to water-based industrial coatings.
A basic condition required for proper substrate wetting is that the surface
tension of the liquid be equal to or lower than the surface tension of the
substrate so that the contact angle measured through the paint is as close to
zero as possible. The concept of "critical surface tension" of a substrate was
developed by Zisman [46], where a liquid with a surface tension higher than
the critical surface of the substrate will not completely wet the substrate (to
give a zero contact angle). Water, with a high surface tension, will therefore
not wet substrates as readily as organic solvent solutions which have low
surface tensions. Water-thinned coatings are therefore more likely to give
poor wetting properties on low-energy surfaces. Nevertheless emulsion paints
contain surfactants used in polymerization, which will lower the surface ten-
sion to enable the wetting of most surfaces such as wood, metals, and most
paint surfaces.
When emulsion paints were first produced and the problem of foaming
was encountered, silicones were tried as antifoams in such systems. These
290 Porter

early silicone antifoams gave excellent antifoam behavior but caused severe
problems in the application of the paint film itself [47-51]. Thus paints
containing silicone can give uneven areas which are often known as fish-
eyes or cissing. These are usually substrate spots where the paint has not
wetted the surface so that holes in the paint film form after drying. The
author has found that other antifoams can also give fish-eyes.
Microscopic examination of fish-eyes usually reveals a particle in the paint
at the center of the eye. The particle is often, but not always, an antifoam
in the paint film which wets out on the substrate and gives an area of the
substrate which aqueous solutions of higher surface tension will not wet.
The antifoam is insoluble in the wet paint [2,47] and will wet out on the
substrate if the surface tension of the antifoam is lower than that of the
critical surface tension of the substrate. Silicone oils and mineral oils can
meet this criteria with many substrates. The particle size of a silicone droplet
is important as large droplets tend to give fish-eyes, while very small sili-
cone droplets will not [48]. The fish-eyes produced by antifoam contami-
nation on a substrate are surprisingly large compared to the size of the
contamination.
A similar phenomenon has been examined by Wilkinson et al. [52] and
Aronson et al. [53J who studied an oil drop on a solid substrate partially
submerged in a thin film of water of varying thickness. Decreasing the thick-
ness of the film of water initially causes no change in the oil drop diameter
while the oil/water /substrate contact angle is increasing from its equilibrium
to its advancing angle; once this is reached, the oil drop spreads rapidly
until, at a critical water-film thickness, it disrupts or separates from the water
phase, to form an oil drop surrounded by bare substrate. The process is
depicted in Fig. 5. It is a possibly analogous to that by which a small oil
drop can give a defect in a paint film. The system studied by Wilkinson et
al. [52] and Aronson et al. [53] involves a homogeneous substrate, and the
situation could become more complex if the substrate itself varies in surface
properties.

B. Loss of Color
The pigments included in emulsion paints to give opacity are usually tita-
nium dioxide (rutile) and cheaper materials such as talc. Other pigments add
color. Such pigments are usually insoluble in water and are contaminated
with surfactant to facilitate dispersal throughout the paint. When a pigment
is added to an opacifying pigment, there can be a redistribution of surfactant
leading to flocculation. The paint can then apparently lose color. It has been
found that antifoams with their associated surfactants can take part in this
process [48,49,51,54-56].
Antifoams for Paints 291

J ::r:
Air

Water Water
(a) Oil
Substrate

(b)

(c)

(d)

FIG. 5 Schematic illustration of dewetting of aqueous film induced by bridging


droplet [52). (a) Formation of configuration. (b) Water height lowered, contact angle
increases to its advancing angle. (c) Oil forced to spread at its advancing contact
angle displacing water. (d) Dewetting of substrate. (From Ref. 52. Reproduced by
permission of Academic Press Inc.)

An antifoam can rob the pigment stain particles of surfactant, and the
pigment or stain can agglomerate, resulting in a change of color usually
going to a paler shade. This process is slow, taking up to 48 h to complete.
The effect of antifoams on the loss of color of a paint formulation should
therefore not be tested for at least 48 h after mixing the paint. Delileo [54]
gives a good account of the practical problems encountered in color stability
and the adverse effects of antifoams.

C. Loss of Gloss
Antifoams can often cause a loss in the gloss of emulsion coatings partic-
ularly if there is a high concentration of hydrocarbon oil in the antifoam.
On matt emulsion paints this is not important but on the newer semigloss
and now high-gloss water-based paints this can be a serious defect and there-
fore must be evaluated in any test procedure.
292 Porter

VI. TESTS FOR ANTIFOAM EFFECTIVENESS


Foam problems in surface coatings occur in manufacture, can filling, and
application. Testing of antifoams should occur under conditions relevant for
each of these processes. Tests for the effectiveness of antifoams under ap-
plication conditions should therefore take into account the possibility of the
loss of antifoam effectiveness on storage (see Kelly et al. [57]. Thus paints
can often be stored for periods of greater than a year before use. During
this time antifoam may adversely interact with other paint ingredients and
may even be slowly solubilized [58].
Evaluation of antifoam in paints can be misleading if the paint already
contains an antifoam. This is invariably the case if the sample is obtained
from a normal large-scale manufactured batch. It is therefore usual to make
paint samples in the laboratory.

A. Paint Manufacture
Two separate stages in paint manufacture can give rise to foam problems.
These are during pigment dispersion at high shear and during addition of
binder to the pigment dispersion at low shear. Laboratory tests for antifoam
efficacy involve air entrainment followed by measurement of the volume of
foam generated. Agitation of the paint with a high-speed stirrer is a common
method of air entrainment [48,58]. Other methods include air injection de-
vices [59-61] and agitation in the standard paint shaker used for mixing
color tints into household paints. Use of a perforated disk which moves up
and down in a cylinder [62] is, however, believed by the author to simulate
manufacturing conditions more closely than air injection, stirring, or shak-
ing. It is shown in Fig. 6. Here it should be stressed that different methods
of air entrainment can give rise to different rankings of the efficiencies of
a series of antifoams.
Quite often 20 or more antifoams will be tested at various concentrations
on a single paint. The apparatus used for testing should therefore be simple,
easy to clean, easy to use, and yield reproducible foam measurements. Ease
of cleaning is particularly important because trace amounts of certain anti-
foams (silicones, for example) can have a significant affect on foam behavior.
In these test methods the foam is readily observed and easily measured
visually by volume in a suitable graduated vessel. Elaborate techniques em-
ploying for example photoelectric devices are seldom justified for the eval-
uation of antifoams for addition at the manufacturing stage. The rate of de-
cay of the foam is sometimes measured by taking observations at times up
to 1 hr [58]. In comparing the effects of different antifoams on foam decay,
it is, however, desirable to compare complete decay curves rather than in-
tegrated foam-time curves as described by Kushchnir et al. [2]. Integration
Antifoams for Paints 293

FIG. 6 Foam generation using perforated disk in cylinder. (From Ref. 62).

can be misleading because foams with radically different decay profiles may
give identical values of the decay integral.

B. Can Filllng
Weight, not volume, is the dominating factor in filling cans of paint. It is
impossible to meet weight specifications if the paint contains dispersed air.
Specific gravity measurement is therefore usually used for evaluating anti-
foams designed to be effective in preventing problems in can filling. It is a
quick and easy measurement [57) which is readily done on those samples
aerated in the various ways described above (Sec. VI.A.). It is particularly
useful when the amount of foam is small and not easy to measure visually
by volume.

C. Application
The most important tests concern examination of the effect of the antifoam
on the properties of the final dry paint film. If the antifoam gives no defects
294 Porter

on the paint film, then it is selected on the basis of the tests described above.
On the other hand if the paint film shows defects which can be attributed
to the antifoam then no matter how efficient the antifoam (as measured by
those tests) it will not be used in the paint. Instead a less efficient antifoam
which gives no defects will be selected.
For household emulsion paints tests on the film forming properties of the
paint simply involve coating the paint by brush and/or roller on a substrate,
drying the paint, and visually examining the dry paint film for defects. Var-
ious workers use different substrates and methods of evaluation [ 1 ,5 ,48 I.
The defects observed are due either to particles of antifoam disrupting the
wetting of the substrate by the coating or to the presence of air bubbles
arising from antifoam inefficiency. An excellent description of these defects
is given by Koerner et al. [48]. Application testing methods for industrial
paints are essentially similar to those for household emulsion paints. How-
ever, industrial paints are often applied from baths which are under recir-
culation. Therefore it is desirable to run a simulated recirculation test to
check antifoam efficiency over prolonged time intervals as described by Neilsen
l42].

D. Stability on Storage
All foam control agents in paint systems lose some efficiency in storage.
Newsome [58] attributes this phenomenon to interaction of surfactants in the
paint with particles of antifoam. Storage tests are often accelerated by using
elevated temperatures of 40-45°C. However, Gress [ 1] suggests storing paints
for four-six weeks at room temperature. Newsome [58], on the other hand,
uses gentle agitation of the paint for periods of about 60 min to simulate
aging. The usual tests of foam volume after air entrainment, specific gravity
and evaluation of surface defects are done on samples of paint before and
after the storage test.

VII. SPECULATIONS ON FUTURE DEVELOPMENT


A. Water-Based Emulsion Paints for Household Use
Emulsion paints will constitute the largest proportion of household paints in
the future. Cost will be the main criterion in deciding the approach to foam
control in such paints. Foam control will therefore not be obtained using
low-foam surfactants unless they become as cheap and efficient as those
which are used at present. Here we note that acetylenic derivatives, fluor-·
inated surfactants, and siloxane copolymers are now being produced which
give excellent emulsifying properties combined with low foam. These low-
foam surfactants could be utilized either directly in the paint and/or in the
Antifoams for Paints 295

manufacture of emulsion polymers and other resin components. Undoubt-


edly research will continue in this area io find surfactants which give the
required surfactant properties as well as keeping foam to a minimum. How-
ever, this type of surfactant is significantly more expensive than those at
present used in emulsion polymers. It seems unlikely then that such surfac-
tants will replace conventional high-foaming surfactants to any significant
extent for the foreseeable future.
Foam control will in the main continue to rely upon the antifoam added
by the paint manufacturer. Low-cost hydrocarbon/ester/soap products will
undoubtedly still be used. Ways may even be found of utilizing hydrophobed
silica/polydimethylsiloxane mixtures without producing coating defects. It
is possible that polyalkoxypolysiloxane siloxane copolymers will become
increasingly utilized in this area, but they will be more often used as ad-
mixtures with other components rather than used alone.
Hydrophobic silica will still be used extensively as an antifoam promoter
for both silicones and hydrocarbons. Improved methods of hydrophobing
will, however, be developed to yield better stability against problems of
sedimentation and chemical interaction.

B. Industrial Paint Systems


The main practical advantage of a low-foam paint system is that application
can be carried out more easily and rapidly. This will mean that cost con-
sideration may sometimes be compromised to secure low-foam behavior.
Similar types of antifoam products to those mentioned above will predom-
inate with the main emphasis on silicone-modified polymers as these have
so far shown most promise. Again they will be utilized in blended systems
rather than used on their own because of lower cost and more versatile
perfom1ance.

REFERENCES
1. W. Gress, Polym. Paint Colour J. 174: 452 (1984).
2. P. Kuschnir, R.R. Eley, and F. Louis Floyd, J. Coat. Tech. 59: 75-87 (1987).
3. R. Heyden, A. Asbeck, M. Eckelt, M. Petzold, and G. lJphues (assigned to
Henkel Kommanditgesellschaft auf Aktien), US 4,094,812: June 13, 1978;
filed December 20, 1976.
4. R. Groves, Industrial Applications of Surfactants (D. R. Karsa, ed. Royal
Society of Chemistry, 1987, p. 73.
5. A. C. Nielsen, J. Water Borne Coat. May: 9 (1980).
6. E. P. McGinn (assigned to Nopco Chemical Company) lJS 2,563,856: August
14, 1951; filed October 12, 1945.
7. E. P. McGinn, (assigned to Nopco Chemical Company) US 2,563,857: August
14, 1951; filed October 12, 1945.
296 Porter

8. C. E. Snook (assigned to Nopco Chemical Company) US 2,715,614; August


16, 1955, filed March 31, 1955.
9. J. H. Shott and J. E. Ward (assigned to Nopco Chemical Company) US
2,849,405: August 26, 1958; filed December 28, 1953.
10. M. Gruenert, H. Hempel and H. Tesmann (assigned to Henkel AG) EP 0036
597: February 15, 1984; filed September 30, 1981.
11. C. C. Currie (assigned to Dow Chemical Co.) US 2,632,736: March 24, 1950;
filed August 22, 1946.
12. F. J. Boylan (assigned to Hercules Powder Company) US 3,076,768: February
5, 1963; filed April 5, 1960.
13. R. Liebling and N. M. Canaris (assigned to Nopco Chemical Company), US
3,207,698: September 21, 1965; filed February 13, 1963.
14. R. E. Sullivan (assigned to Dow Coming) US 3,383,327: May 14, 1968.
15. Assigned to Witco Corporation GB 1 490 393; filed January 9, 1975.
16. F. J. Boylan, C. Porter, and 0. D. Bruno (assigned to Drew Chemical Corp.)
US 4,377,493: March 22, 1983; filed June 26, 1981.
17. Assigned to Shinetsu Chemical Company, GB 1 468 896; March 30, 1977;
filed June 7, 1974.
18. W. J. Raleigh (assigned to General Electric Company) US 4,005,044: January
25, 1977; filed June 20, 1975.
19. I. A. Lichtman and F. E. Woodward (assigned to Diamond Shamrock) US
3,697,440: October 10, 1973; filed June 27, 1969.
20. F. J. Boylan (assigned to Hercules Incorp.) US 3,705,859: December 12, 1972;
filed September 23, 1970.
21. H.J. S. Shane, J.E. Schill, J. W. Lilley (assigned to Hart Chemical Company
Ltd., Canada) US 3,723,342: March 27, 1973; filed December 21, 1971.
22. D. W. Suwala (assigned to Diamond Shamrock Corp.) GB 1 480 717: July
20, 1977; filed November 3, 1975.
23. J. H. Curtis and F. E. Woodward (assigned to Diamond Shamrock) US
3,673,105: June 27, 1972; filed April 11, 1969.
24. R. Heyden, A. Asbeck, and E. Echelt (assigned to Henkel KGA) US 4,087 ,398:
June 2, 1978; filed November 29, 1976.
25. R. Heyden and M. Eckett (assigned to Henkel and Cie GmbH), US 3,919,911:
November 11, 1975; filed February 28, 1973.
26. J. V. Sinka and I. A. Lichtman (assigned to Diamond Shamrock Corp.), US
4,021,365: May 3, 1977; filed August 18, 1975.
27. R. J. Michalski and R. W. Youngs (assigned to Nalco Chemical Co.), US
3,923,683; December 2, 1975; filed January 1, 1974.
28. F. J. Ihde (assigned to Diamond Shamrock Corporation) US 4,123,383: Oc-
tober 31, 1978; filed September 19, 1977.
29. I. A. Lichtmann and A. M. Rosengart (assigned to Diamond Shamrock Corp.),
US 3,677,963: July 18, 1972; filed September 30, 1970.
30. R. W. Kaychok and F. J. Boylan (assigned to Drew Chemical Corp.), US
4,626,377: December 2, 1986; filed September 10, 1984.
31. W. T. Flannigan (assigned to Imperial Chemical Industries PLC), EP 76 558:
April 13, 1983; filed June 7, 1982.
Antifoams for Paints 297

32. J. C. Wuhrmann, H. Mueller, K. D. Brands, A. Asbeck, and J. Heidrich


(assigned to Henkel Kga) EP 147 726: December 12, 1984; filed January 2,
1984.
33. W. T. Schmidt and C. T. Gammon (assigned to Diamond Shamrock Corpo-
ration) US 4,225,456: September 30, 1980; filed November 6, 1978.
34. Assigned to Badische Anilin- and Soda-Fabrik Aktiengeselschaft GB 1,402,597:
August 13, 1975; filed November 17, 1971.
35. W. J. Raleigh (assigned to General Electric Co.), US 4,005,004: January 25,
1977; filed July 20, 1975.
36. C. T. Gammon (assigned to Diamond Shamrock Corp.), US 4,208,301: June
17, 1980; filed July 7, 1978.
37. C. Manohar and V. K. Kelkar, J. Colloid lnteiface Sci. 137: 604 (1990).
38. H. Steinberger and C. Werner (assigned to Bayer Aktiengesellscaft), US
4,562,223: December 31, 1985; filed August 4, 1983.
39. H. F. Fink, 0. Kloeker, G. Koerner, G. Rossmy, and C. Weitemeir (assigned
to Th. Goldschmidt AG) US 4,520,173: June 28, 1985; filed April 7, 1983.
40. J. Lucassen, Faraday Disc. Chem. Soc. 59: 76 (1975).
41. M. J. Schnall, Proc. lOth Water Borne and High Solids Coating Symp., Lou-
isiana, 1983, p. 46.
42. A. C. Nielsen, J. Water Borne Coat. Feb: 17 (1980).
43. Basildon Chemicals, private communication.
44. R. Berger, H. F. Fink, and 0. Kloeker (assigned to Th. Goldschmidt AG),
EP 0 257 356: March 2, 1988; filed August 3, 1987.
45. H. U. Hempel, G. Grunert, H. Tesmann, and H. Muller (assigned to Henkel
KGA) US 4,504,410: March 12, 1985; filed June 16, 1982.
46. W. A. Zisman, J. Paint Technol. 44(6):41 (1972).
47. M.A. Ott, Mod. Paint Coat. Aug: 31 (1977).
48. G. Koerner, F. Fink, R. Berger, and W. Heilen, World Surfactant Congress,
IV, 211 (1984).
49. R. D. Nicholson, Austral. OCCA Proc. News Aug: 9 (1983).
50. R. T. Maher and E. M. Antonucci, Mod. Paint Coat. Dec: 43 (1979).
51. L. Kelley Mod. Paint Coat. June: 33 (1975).
52. M. C. Wilkinson, A. C. Zettlemoyer, M. P. Aronson, and J. W. Vanderhoo,
J. Colloid lnteiface Sci. 68(3): 508 (1979).
53. M. P. Aronson, A. C. Zettlemoyer, and M. C. Wilkinson, J. Phys. Chem.
77(3): 318 (1973).
54. L. M. Dilelio, Resin Rev. 20: 29 (1977).
55. E. M. Antonucci, Water Borne and Higher Solids Coatings Symp. Feb. 12-
14 New Orleans, LA, 1979 p. 234.
56. E. M. Antonucci, P. Cosetino, R. W. Stangs, and T. F. O'Farrell, Water
Borne Coat. Nov: 12 (1977).
57. E. L. Kelley, R. W. Harrison, and B. B. Harris, Polym. Paint Colour J. Aug:
734 (1978).
58. G. Newsome, Paint and Resin, March/April (1982).
59. Deutsches Institut fur Normung DIN 51381.
60. American Society for Testing and Materials ASTMD -892-72.
61. Deutsches lnstitut fur Normung DIN 51566.62. Deutsches Institut fur Nor-
mung DIN 53902 Tei! I.
8
Surfactant Antifoams

TREVOR G. BLEASE, J. G. EVANS, AND L. HUGHES ICI


Surfactants, Wilton, Middlesbrough, United Kingdom
PHILIPPE LOLL ICI Surfactants, Everberg, Belgium

I. Introduction 300
II. Types of Polyoxyalkylene Surfactant Antifoam 300
A. Polyoxyethylene/polyoxypropylene block copolymers 300
B. Ethylene diamine-based polyoxyethylene/polyoxypropylene
block copolymers 30 I
C. Polyol-based polyoxyalkylenes 301
D. Fatty alcohol-based polyoxyalkylenes 302
E. Fatty alcohol polyoxyalkylene alkyl ethers 302
III. Ecotoxicological Aspects of Copolymer Surfactant Antifoams 303
A. Biodegradation of copolymer surfactant antifoams 303
B. Aquatic toxicity of antifoams 305
IV. The Use of Copolymer Antifoams in Machine Dishwashing 305
A. Machine types and operating conditions 306
B. Selection of surfactant anti foams 307
C. The defoaming performance of polyoxyethylene/polyoxy-
propylene copolymers under dynamic conditions 309
V. Use of Copolymer Antifoams in Sugar Beet Industry 311
A. Introduction 31 l
B. Description of a sugar plant 314
C. Causes of foam in the sugar beet industry 317
D. Composition of the antifoam 318
E. Toxicological implications 321
References 321

299
300 Blease et al.

I. INTRODUCTION
Other chapters in this volume have described some important industrial
applications of antifoams. This contribution focuses on two particular ap-
plications, namely dishwashing and sugar beet processing, for which poly-
oxyethylene/polyoxypropylene-derived antifoams have played an important
role in satisfying the foam control demands of the industry. For dishwashing
a distinction will be made between domestic usage and industrial/institu-
tional application due to their differing performance requirements. Ecotox-
icological issues surrounding the polyoxyalkylene copolymer products have
been an important catalyst for new product development in this area over
recent years and these matters will be covered in a separate section.
It is important to note at the outset that the antifoaming power of the
surface-active agents covered in this chapter is temperature dependant. Above
the cloud point of the surfactant in the application medium (the temperature
at which hydrogen bonding is insufficient to maintain the surfactant in so-
lution) the antifoaming power is considerably enhanced [1].

II. TYPES OF POLYOXYALKYLENE SURFACTANT


ANTI FOAM
In the application areas here described only a few classes of polyoxyalkylene
surfactant are commercially significant. For this chapter a relatively short
description, paying particular regard to the antifoaming aspects of the struc-
tures, will suffice. A more general treatment of the compound classes and
their properties can be obtained from other texts [2,3].

A. Polyoxyethylene /Polyoxypropylene Block


Copolymers
These products, simply consisting of polyoxyethylene and polyoxypropylene
blocks, were first commercialized in the early fifties by Wyandotte Chem-
icals Corporation [4]. The generic structure is typically
HO(EO>a(POMEO)aH,
where
EO = -CH2CH20- ethylene oxide group
PO = -CH2CH(CH3)0- propylene oxide group
Their utility is evidenced by their widespread use for industrial antifoaming
to this day. Both "normal" (above) and "reverse" block structures, that is,
Surfactant Antifoams 301

FIG. 1 Structure of ethylene diamine-based polyoxyethylene polyoxypropylene block


copolymer.

are employed. The terminal secondary hydroxyl group of the reverse block
structure offers some chemical stability benefits compared to the primary
hydroxyl functionality of the normal block products. Optimal antifoam ef-
ficiencies, for both structural types, are obtained at polyoxypropylene: poly-
oxyethylene molar ratios between 4: 1 and 9: 1 and molecular weights greater
than about 2000 amu [2]. A product for which a = 2 and b = 32, with a
cloud point in 10% aqueous solution of about l 7°C, would be considered
typical of the normal block antifoam products available from several man-
ufacturers. By varying the molar ratio a continuum of cloud points up to
greater than 100°C can be obtained.

B. Ethylene Diamine-Based Polyoxyethylene/


Polyoxypropylene Block Copolymers
These products have been available in normal (hydrophile end-block) and
"reverse" (hydrophobe end-block) structures, as illustrated in Fig. l, since
the 1950s. For useful antifoaming performance, molecular weights are
somewhat higher than for the polyoxyethylene/polyoxypropylene block co-
polymers, typically greater than 3500 amu [2]. As for the products described
in Sec. A, the cloud point can be continuously varied from 10°C to greater
than 100°C.

C. Polyol-Based Polyoxyalkylenes
Many polyols are employed as starting bases for a wide variety of more
specialised polyoxyalkylene antifoams. Among the polyols most commonly
employed presently are
Glycerol
Sorbitol
Pentaerythritol
Trimethylolpropane
Such products can yield relatively better antifoam performance than con-
ventional polyoxyethylene/polyoxypropylene block copolymers. Moreover,
302 Blease et al.

polyol-based products can be used as antifoams for applications in which


multifunctionality is desired. Multifunctionality can be achieved by various
partial postreactions of the polyolpolyoxyalkylene.

D. Fatty Alcohol-Based Polyoxyalkylenes


This class of polyoxyalkylene surfactant antifoams, of generic structure
RO(EO)x(AO)yH
where
R = C 8_ 18 linear (or substantially linear) alkyl chain
EO = ethylene oxide group
AO = propylene oxide or butylene oxide group,
is assuming increasing importance nowadays. The awareness of the en-
hanced biodegradability possibilities of these surfactant antifoams compared
to other compound classes (see Sec. III) has led to replacement of the tra-
ditional polyoxyalkylene antifoam products by fatty alcohol-based polyoxy-
alkylenes in many important applications, for example dishwashing (see Sec.
IV).
In order to obtain the optimum biodegradability performance these mol-
ecules must be kept small, that is x and y generally in single figures [5].
Within this relatively limited scope, seemingly small but highly significant
structural variations must be played, involving, for example,
The choice of fatty alcohol
The numbers of polyoxyalkylene moles added
The internal arrangement of the polyoxyalkylene moles (block, normal or
reverse, or random)
The choice of propyleneoxide or butylene oxide as the hydrophobic
polyoxyalkylene
in order to best meet the specific antifoaming performance requirements of
the application.

E. Fatty Alcohol Polyoxyalkylene Alkyl Ethers


Evolving from the fatty alcohol-based polyoxyalkylene antifoams in the 1980s
have been the etherified, so called end-capped, derivatives [6,7]. These mol-
ecules are of generic structure:
RO(EO)x(AO)yR'
where R, EO, and AO are already defined and R' is generally a small (sC 4 )
linear alkyl chain.
Surfactant Antifoams 303

The acquisition of highly biodegradable hydrophobicity, via the addition


of the linear alkyl end cap, can allow reduction or even complete replace-
ment of the hydrophobic polyoxyalkylene component (i.e., y tends to zero)
for the equivalent antifoaming performance. Additionally the substitution of
an ether linkage for the chemically comparatively labile hydroxyl function-
ality is advantageous for industrial applications involving prolonged expo-
sure to stringent chemical conditions, most notably the extreme pH envi-
ronment of segments of the industrial and institutional detergency market.

Ill. ECOTOXICOLOGICAL ASPECTS OF


COPOLYMER SURFACTANT ANTIFOAMS
A. Biodegradation of Copolymer Surfactant
Antifoams
1. Polyoxyethylene/Polyoxypropylene Block Copolymers
Amongst the earliest surfactant antifoams were the ethylene oxide/propyl-
ene oxide copolymers end blocked with either the polyoxyethylene hydro-
phile or the polyoxypropylene hydrophobe. When used as antifoams these
copolymers have molecular weights typically greater than 2000 amu. They
are poorly biodegradable because the polyoxypropylene moiety confers a
high degree of bioresistance [8].
The ease of biological degradation of these antifoam copolymers is de-
termined by their structure. For ethylene oxide/propylene oxide copoly-
mers, both the hydrophylic (polyoxyethylene) and hydrophobic (polyoxy-
propylene) entities influence the degree and extent of degradation.
Oxypropylene groups in these copolymers have a negative effect on rate and
degree of biodegradation. Poor degradability of oxypropylene chains is usu-
ally associated with restriction of the hydrolytic sequence due to steric hind-
erance by pendant methyl groups. Other factors such as the lower solubility
of the polyoxypropylene moiety compared to polyoxyethylene may further
limit the bioavailability of these structures to microorganisms.
The hydrophylic polyoxyethylene chain on the otherhand degrades more
readily but at a rate dependent upon molecular weight. Thus ethylene glycol
and its oligomers have been shown to exhibit biological degradation in lab-
oratory models [9]. Several polyoxyethylene degrading bacteria capable of
growing on ethylene glycol with chain lengths in excess of 1OOO amu have
been isolated from the environment 19]. While polyoxyethylene chains in
excess of 1000 amu are considered to be less readily biodegradable, eventual
aerobic degradation may occur [ 1O]. Anaerobic degradation of polyoxy-
ethylene by microorganisms has also been reported [ 11]. Polyoxyethylene
chain degradation in ethylene oxide/propylene oxide copolymers is consid-
304 Blease et al.

ered to occur by initial cleavage at the a carbon of the hydrophobe. This is


followed by sequential hydrolytic, and oxidative breakdown of the end gly-
col groups by the (3-oxidation pathway [ 12]. Biodegradation of unsubstituted
polyethylene glycols is reviewed in detail by Swisher [13].
Legislation in particular within the European Community has limited the
application of ethylene oxide/propylene oxide antifoams for certain deter-
gent applications. Failure to meet the legal requirement of a minimum 80%
primary biodegradation for this class of compounds has led to the search for
other antifoam structures which do. Typically, the degree of primary bio-
degradation which usually represents a limited breakdown of the substance
but not total demineralization to water and carbon dioxide in standard screening
and confirmatory tests is of the order of 10%-30% for ethylene oxide/pro-
pylene oxide block copolymers [14]. Primary biodegradation is a measure
of the removal of the molecule from the aqueous phase in contrast to ulti-
mate biodegradation-the complete mineralization to inorganic products such
as carbon dioxide and water.
The importance of structure in determining the biodegradation of surfac-
tant antifoams is well understood. Much attention has centred on the hy-
drophobic character of the propylene oxide group, the total carbon chain
length, chain length distribution, branching and substitution. The hydro-
philic ethylene oxide portion, although less influential, may become a de-
termining factor particularly in relation to chain length.

2. Fatty Alcohol Polyoxyethylene /Polyoxypropylene


Derivatives
The development of fatty alcohol alkoxylates for antifoam applications has
centered on the C 8 to C 15 fatty alcohol hydrophobes. In most cases, the
length and type of fatty alcohol has only a minimal effect on biodegradability
unless a high degree of branching is present [14]. Propylene oxide groups
have a dominating effect on the degree of biodegradability compared to the
influence of the corresponding ethylene oxide groups [14]. Limiting values
for propylene oxide of between 7 and 10 units within alcohol alkoxylates
for compliance with European Community primary biodegradation require-
ments have been reported [5].
Other studies of primary and ultimate biodegradability using semicontin-
uous activated sludge (SCAS) bioreactors revealed that the insertion of pro-
pylene oxide groups into the polyoxyethylene chain of fatty alcohol ethoxyl-
ates decreases the rate and extent of biodegradation. The magnitude of the
effects were proportional to the propylene oxide block size and the extent
of branching in the alcohol [15]. Alcohol alkoxylate antifoams based on a
C 13 -C 15 oxoalcohol with varying polyoxypropylene and polyoxyethylene
groups assessed in standard tests [ 16) show moderate ultimate degradation
Surfactant Antifoams 305

TABLE 1 Polyoxypropylene and Polyoxyethylene Block Copolymers Acute Fish


Toxicity

Rainbow trout
Molecular 96 hr LC50
weight (mg/L)

Pol yox yethylene( 5) Polyoxypropylene(32) 2090 203


Polyoxyethylene( I 0) Pol yox ypropy lene(3 2) 2400 >1000
Polyoxyethylene(25) Polyoxypropylene(32) 2900 315
Polyoxyethylene(l 52) Polyoxypropylene(32) 8350 >1000
Polyoxyethylene( 16) Polyoxypropylene( 48) 3450 >1000
Note: Figures in parentheses refer to number of moles of oxypropylene or oxyethylene.
Source: Ref. 17.

after 28 days [ 17]. This compares with block copolymer antifoams which
have a low degree of ultimate biodegradability in the same tests.
3. Fatty Alcohol Polyoxyethylene Alkyl Ethers
Defoaming agents based on fatty alcohol polyoxyethylene alkyl ethers are
reported [18) to give excellent primary (98%) and ultimate biodegradation
(75%) in standard screening tests [ 16].
B. Aquatic Toxicity of Antifoams
It is well known that polyoxyethylene/polyoxypropylene block copolymers
are considered to be low-risk compounds for impact on aquatic life. Typi-
cally, the degree of acute aquatic toxicity, determined as the 96-hr LC 50
value (the chemical concentration lethal to 50% of fish exposed over 96 hr),
is greater than 100 mg/L [14]. Studies on a range of polyoxypropylene/
polyoxyethylene block copolymers for acute toxicity to rainbow trout have
confirmed the low toxicity of this antifoam class (see Table 1). By contrast
fatty alcohol polyoxyethylene polyoxypropylene compounds introduced as
alternatives to the ethylene oxide/propylene oxide block copolymers for bio-
degradation reasons display LC 50 values of 0.7-2.9 mg/L toward a number
of freshwater fish species [ 17). The toxicity to freshwater invertebrates, par-
ticularly Daphnia magna, is found to be comparable to the toxicity range of
0.25 to 4.4 mg/L expected for typical alcohol ethoxylates [17]. Thus for
these fatty alcohol derivatives improved biodegradability with respect to block
copolymers is associated with increased aquatic toxicity.

IV. THE USE OF COPOl YMER ANTIFOAMS IN


MACHINE DISHWASHING
The aim of automatic dishwashing is the complete cleaning and drying of
crockery and utensils, leaving them ready for immediate use. This is achieved
306 Blease et al.

by manual removal of large food particles prior to loading in to the machine,


followed by freshwater prewashing, washing, rinsing, and drying cycles.
Both batch and continuous dishwashing machines are used. One common
feature of automatic dishwashing, using either type of machine, is the rel-
atively higher energy levels compared to manual operation. Foam is per-
ceived as a benefit in the latter operation for aesthetic reasons and as an
indication of effectiveness. By contrast the efficiency of circulation systems
in automatic machines is significantly reduced when foam is present. Thus
the surfactant requirements for the low-foam operation necessary in machine
dishwashing are very different to those of hand dishwashing.
Foam generation during automatic dishwashing is influenced by several
factors, including the machine type, the washing program selected, and the
type and level of soils present.
Degradation of proteinaceous soils such as dairy products under the con-
ditions found in the washing cycle (high temperature, high pH) leads to the
formation of surfactants and to the generation of unacceptable levels of foam.
Increased use of surfactants as emulsifiers in the preparation of convenience
foods also contributes to the problem.

A. Machine Types and Operating Conditions


1. Batch Machines
There are many types of domestic and small-scale industrial batch dish-
washing machines. The ware to be washed is placed in a tray or rack which
is loaded into the dishwasher. It remains static during the cleaning operation
which is achieved using rotary spray arms. A typical machine is illustrated
in Fig. 2. Typical operating conditions arc given in Table 2.
2. Continuous Machines
This type of machine is illustrated in Fig. 3. It differs from the batch equip-
ment previously described in several ways:
l. Ware is loaded on to a moving rack which carries it through the suc-
cessive cleaning and drying cycles.
2. Throughput times are lower (less than l min).
3. Capacities are much higher, e.g., over 7000 plates per hour.
4. Pump rates and pressures are significantly higher than batch operated
systems (2.5-3.0 bar).
As with batch industrial machines the final rinse water is recycled back into
the prcwash and wash tanks to conserve energy. This introduces residual
rinse aid into the washing cycle. The nett result is a greater demand on the
defoaming performance of the surfactants used.
Surfactant Antifoams 307

Removable siding trays


for crockery, cutlery,
utensils etc.

FILTERS, RECIRCULATION CONTROL


PUMP, HEATER, WATER PANEL
INLET/EXIT

FIG. 2 Schematic of batch dishwasher.

B. Selection of Surfactant Antifoams


Surfactant foam control in the automatic dishwashing process is achieved
by the incorporation of surfactant antifoams into the rinse aid and/or the
main wash detergent. Levels of surfactant in main wash detergents are typ-
ically :::: 5%, whereas in rinse aids they are considerably higher, typically
16-25% by weight.
Careful selection of the surfactant antifoam(s) is necessary to achieve a
balance between the defoaming characteristics of surfactant with low aqueous

TABLE 2 Typical Operating Conditions for Domestic and Industrial Autodishwashers

Domestic Industrial

Cycle times 60-80 min Typically 1-3 min (up to


30 on some machines)
Temperatures (°C)
Prewashing Ambient 40
Washing 55-65 55-65
Rinsing 60-70 80-85
Drying Up to 90°C Optional
Recycle of water? No Yes
Spray pump pressures Up to 1 bar 1-2 bar
308 Blease et al.

LOAD PRE-WASH WASH' RINSE 2 DRY DISCHARGE


30-40°C 55-65°C 80-85°C 60°C+

Rinse aid
added
Detergent Fresh Water

Fixed Spray
arms - 111111'1

Pre-Wash Tank
Conveyor Belt

Main Wash Rinse Tank



Control Panel
and Pump Tank & Pump & Pump

FIG. 3 Schematic of continuous dishwasher.

cloud points (i.e., less than 30°C) and the better wetting performance ob-
tained with products of higher aqueous cloud points. This is particularly
important for the operation of continuous machines, wherein a surfactant or
surfactant combination must wet effectively at 80-85°C in the rinsing sec-
tion and also perform satisfactorily as a defoamer in the washing section at
around 60°C-two opposing requirements.
Examples of typical formulations are given in Tables 3 and 4 [19-23].
These tables reveal the type of components likely to be found in automatic
dishwashing formulations. The permutations of these and other components
of dishwashing formulations are almost endless and therefore cannot be dealt
with here.
Historically, low-foaming performance has been achieved by the use of
polyoxyethylene/polyoxypropylene copolymers. Key features of these mol-
ecules are their relatively high molecular weights (up to 2500 amu) and high
polypropylene oxide content, which can be up to 90% by weight of the
molecule. This combination allows them to perform efficiently as antifoams
and wetting aids.
However with the introduction of European Community legislation [241
which was due to take effect at the end of 1989, radical changes to molecular
compositions have had to take place for this market, since most of the older
products are not sufficiently biodegradable. As a consequence, at the time
of writing much product and formulation development work is still being
Surfactant Antlfoams 309

TABLE 3 Formulations for Domestic Dishwashing

Powder %w/w Rinse aid %w/w


Nonionic surfactant 3.0 Nonionic surfactant 25.0
anti foam antifoam
Ancillary components 30.0
(hydrotropes, water
softeners, etc.)
Inorganic salts, including 95.0 Demineralized water Balance
sodium metasilicate, so- preservative
dium tripolyphosphate,
sodium carbonate, so-
dium sulfate,
Bleach (chlorine-releasing 2.0
agent)
This type of formulation is to be found in many patents (see, for example [19-23]).

carried out aimed at providing the necessary low-foaming (and wetting) per-
formance while still complying with the biodegradability requirements.

C. The Defoamlng Performance of Polyoxyethylene /


Polyoxypropylene Copolymers under Dynamic
Conditions
Several laboratory methods for the assessment of the foaming tendency of
surfactant systems are available. However, to best simulate the dynamic
conditions found in dishwashing machines, the Association Francaise de
Normalisation method (AFNOR) [25] (Chute et Recyclage) has been used.

TABLE 4 Formulation for Industrial Dish washing

Powder %w/w Rinse aid %w/w

Nonionic 2-5 Nonionic 30.0


surfactant antifoam surfactant antifoam
Inorganic salts 50-55 Ancillary compo- 30.0
(as domestic) nents (hydrotropes,
water softeners, etc.)
Sodium or potassium 15-20 Demineralized water/ Balance
hydroxide preservative
Chlorinating agent/ Balance
fillers/ water
An example of an industrial powder is given in Ref. 23 and an industrial rinse aid in Ref. 22
310 Blease et al.

TABLE 5 Defoaming Performance of Polyoxyethylene/Polyoxypropylene


Copolymers

Biode- Foam height (mm)


Cloud gradable
30°C 50°C 70°C
Product point (°C) [24)

No defoamer present >600 >600 >600


Polyoxyethylene/polyoxy- 22a No 479 805 68
propylene Block copoly-
mer 20% ethylene oxide
Ethylene diamine-based 36. No 328 38 43
polyoxyethylene/polyox-
ypropylene copolymer
Fatty alcohol-based po- 25• No 77 42 46
lyoxyethy Jene /polyoxy-
propylene copolymer
" 40• No >600 208 66
<5.(36b) Yes 50 58 60
25"(48h) Yes 366 80 85
34"(56b) Yes >600 425 90
11 with terminal alkyl group 28" Yes 510 58 65
11 with terminal alkyl group 4lb Yes 116 64 55
'Cloud point measured as I% w/w aqueous solution.
bCloud point measured as 10% w/w in 25% w/w butyl diglycol.
Source: Ref. 17.

The apparatus consists basically of a water-jacketed glass tube and recir-


culating pump. Surfactant antifoam solution is pumped to the top of the
column and allowed to drop freely to the bottom.
The method has been used to screen a number of candidate antifoams and
examples of results are presented in Table 5. Evaluation is carried out in an
alkaline solution containing milk powder as the soil to simulate the cleaning
operation. These results demonstrate
I. The necessity and effectiveness of an antifoam in the dishwash operation.
2. The measure of success achieved in developing sufficiently biodegrad-
able fatty alcohol-based molecules which meet the low-foaming
requirements.
3. The need to select products with the correct cloud point so that a balance
between defoaming and wetting performance can be obtained. In ex-
tremis, a product with a low aqueous cloud point ( < l0°C) will work
well as a defoamer but will give very poor wetting performance, and
vice versa when the product's cloud point exceeds 30-35°C.
Surfactant Antifoams 311

The effect of operating temperature on antifoaming efficiency can be de-


termined by measuring the speed of rotation of a dishwashing machine spray
arm under clean conditions (pure water) and in the presence of proteinaceous
soils. The efficiency can be calculated as follows:
speed of rotation (soil conditions)
Efficiency = x 100%
speed of rotation (water)
Typical results for a series of fatty alcohol alkoxylates, in which the cloud
point is altered by changing the ethylene oxide/propylene oxide ratio can
be seen in Fig. 4 [17, 26).

V. USE OF COPOLYMER ANTIFOAMS IN SUGAR


BEET INDUSTRY
A. Introduction
In contrast to cane sugar manufacture, beet sugar factories need large quan-
tities of antifoams. The natural foaming tendency of aqueous sugar solu-
tions, combined with the high pressures and/or high flowrates, otherwise
leads to unacceptable foaming and overflowing. Engineering modifications
can decrease the foaming at some stages of the process, but in general the
use of antifoaming agents is necessary. The selection of appropriate anti-
foams is a complex matter since a product added at one point in the process
can have effects downstream. This complexity has led to a few specialized
companies, with considerable expertise and experience, supplying antifoam
products to the sugar beet industry.
Antifoams are required at different stages of the process, but the most
important are (I) the beet transport and washing (2) the diffusion (sugar
extraction) (3) postdiffusion (see Fig. 5). The antifoams employed for (1)
arc completely different from those for (2) and (3).
Consumption of antifoams varies from site to site and year to year and
is also greatly dependent on the type of antifoam employed. It has signifi-
cantly decreased over the last decade due to extensive use of more sophis-
ticated pumping devices and more efficient products (vide infra). Despite
certain papers mentioning relatively high consumptions [27, 28, 29, 30] a
rough average total process consumption with efficient modern antifoams
can be estimated at 100 g/tonne of sugar beets. This assumption leads to a
Western European market of some 11,000 tonnes of anti foams.
Many of these antifoams are proprietary mixtures, and therefore the exact
composition is not known. However they are usually complex blends of
several components having different functions [31]. Some of them exhibit
synergistic antifoam effects. Other components have in fact no antifoam
312 Blease et al.

Antifoaming Efficiency (50°C)

>
u
c

...
Ill
·u
;;::
80

\ ...
-
Ill

.:
DI ~
E
......
ftl
0 60


c
<(
...
40
0 20 40 60 80 100
Cloud Point (°CI

Antifoaming Efficiency (80°C)


100

~
c
Ill 80
'y
~
.:DI -~
E 60
......
ftl
0

c
<(

Cloud Point (°C)

Cloud point measured as 1% w/v solution in


demineralised water

FIG. 4 Fatty alcohol alkoxylates defoaming efficiency variation with cloud point
and temperature (cloud points altered by changing ethylene oxide/propylene oxide
ratio). (From Refs. 17 and 26.)
.l.:.'..),sj-.,.

/i'~f,F~
1
Water fJ)
c
(2-3 bar 3 to 6 m 3 /Te of Beet "C I " __.,..
;.
-
Beets
!ear"
Water
~
&
::ll

11-71
Rinsing & Washing I Fresh )lo
:::I
Water
400
I Slicing ..
"Cossettes / , Diffusion
I" Hot Water ~3
UJ
Pulp
Pulp Pressing 1----~

tgz~~~
"'..:··L
460Te/hr
eao Pulp
I

Evaporation
(Syrup 30% water) co, CaC03 so, t17',L;.,_,

-, - \1
,Aff.:' ' & Impurities '~'i.;F;. Alcohol ·;'Alf:.
Vacuum Pan
I
:··:>·-"'· ~~).. /\:s,\·

Fermentation
Crystallising Without 0 2 Distillation
& Molasses
Centrifuging

Sugar
Sold Outside
--fj
·I Fermentation
WithO,
Yeasts
or
Aminated

....ww
Acids

FIG. 5 Schematic of sugar beet processing, highlighting antifoam (A/F) application areas.
314 Blease et al.

TABLE 6 Percentages of Ground Stones Adhering to the Beets Delivered to the


Sugar Plants

% soil + stones
(of total weight)

Dry campaign 8-18


Normal campaign 15-25
Rainy campaign 30-50

function. They serve, for example, as diluents, pourpoint depressants or


spreading agents. However major antifoam components in most sugar beet
antifoam mixtures are polyoxyethylene/polyoxypropylene copolymers.

B. Description of a Sugar Plant [32-34]


A typical flow diagram of a sugar plant is shown in Fig. 5. Stages in the
process where antifoam addition is usually necessary are indicated in the
figure.
1. Washing of the Beets
Sugar beets are crops harvested in most of the United States, Canada, and
Europe between September and November. On delivery to the sugar beet
factory, significant amounts of soil and stones can adhere to the beets, as
illustrated in Table 6.
a. Transportation. At the sugar plant, transport of the beets from the
storage area to the washing area is increasingly performed by a mechanical
process (conveyor belt). However, hydraulic transportation is still dominant.
In this latter case, water guns project recycled water under pressure onto
the beets to transport them in the hopper. Water flowrates of 3 to 6 m3 /t of
beets are necessary to transport about 250 tonnes/hr of beets in 60-cm-wide
hoppers. Typically a sugar plant handles 10,000 tonnes/day of beets on
which 30,000 to 60,000 m3 /day of water are sprayed at a pressure of 2-3
bar. When hydraulic transportation is used, there can be a lot of foam in
the hoppers after the water guns. Thus an antifoam is added to the water
supply of these guns.
The replacement of hoppers by a conveyor belt will obviously affect the
need for antifoams. However, in such a case the washing section must have
a bigger capacity as the hydraulic transportation also contributes to the washing.
b. Washing and Rinsing. Several pieces of equipment permit the removal
of the earth, stones and vegetation to obtain washed beets. Basically clean
water is introduced into the last rinse and flows to the proceeding stage so
Surfactant Antifoams 315

as to be in countercurrent with the circulation of the beets. As there exist


several potential foaming problems during this washing/rinsing process, an
antifoam is added into the clean water tank. This is a preventive action as
this water is used upstream. However, it is sometimes insufficient and some
antifoam must be added to avoid foaming at some specific parts of the circuit.
Once the beets are washed, they are cut into slices (2-3 mm thick,
10-15 cm long) called cossettes.
2. Sugar Extraction: The Diffusion
Sugar is extracted by a continuous countercurrent liquid-solid extraction per-
formed in a diffuser with hot water at 75°C maximum. Essentially the cas-
settes enter at one end of a very large, long cylinder, called the diffuser,
while the hot water enters at the other end. The sugar juice which emanates
from the diffuser is basically a 15% sugar solution in water, although some
impurities are present at this stage. Many different types of diffuser, now-
adays universally known by the names or initials of the original manufac-
turers, are used in the industry (e.g., ODS, Olier, BMA, Buckau, De Smedt,
and RT) differing in their countercurrent geometry. To illustrate the appli-
cation of antifoam agents in the diffusion stage one type of diffuser, the RT
diffuser, will be described in some detail.
a. RT Diffuser. A representative flowchart of a RT diffuser is shown in
Fig. 6. A lot of diffusers are now fitted with a heat exchanger which will
use the heat from the hot sugar juice to prewarm the cassettes.
As the diffuser rotates, large quantities of juice are falling in the tank.
At this stage, due to the relatively high temperatures involved (70-75°C)
and to the mechanism of action of the copolymers (see Sec. I), it is not too
difficult to obtain a low level of foam.
It is, however, more difficult to obtain the same result with the cold juice
falling from the heat exchanger because its temperature is lower. Thus higher
quantities of antifoam are required. It is often preferred to use the same
antifoam for both temperatures, and the choice of the right copolymer(s) is
of importance.
The antifoams are usually added to the hot juice and cold juice tanks.
Some products give better results when directly added onto the cold cas-
settes. Small amounts may also be introduced in the hot water coming from
the pulp processing.
In total, an RT diffuser requires about 20-40g of an efficient antifoam
mixture/tonne of beets.
b. BMA and Buckau WolfTower Diffusers. In these diffusers an antifoam
is mainly needed in the cossette mixer. Cossettes are mixed with the hot
juice transferred from the diffuser, which in this case is a vertical tower, by
a pump. The antifoam will also help the draining of the juice through the
316 Blease et al.

400 Ta/hr Hot Water


"Cosaettas"
s•c I

Diffuser
IRT)
Hot Sugar Pulp
Juice 460 Ta/hr
Water

Pulp Pra11ing

Pulp

FIG. 6 Schematic of RT sugar beet diffuser.

perforated disks and thus reduce the power consumption. A BMA diffuser
would require 10-20g of antifoam/tonne of beets.
c. Desmet Diffuser. This diffuser consists of a 33-m-long conveyor screens
which moves a 0.6- to 1-m thick layer of cossettes below a series of juice
sprayers. Basically each sprayer is fed with the juice collected after the pre-
ceding one. The foaming is particularly a problem at the exit of the diffuser
where the juice is most concentrated. Thus the antifoam is often added in
the last part of the diffusion and it is common to need 60 g/t of beets.
3. Purification and Crystallization
The sugar juice is screened and then a preliming occurs. The lime used is
often in the form of an aqueous solution known as milk of lime. The next
step is the addition of the balance of liming agent and continuous bubbling
of carbon dioxide. This will precipitate the lime as small insoluble calcium
carbonate crystals on which most of the impurities will adhere. This oper-
ation is called carbonatation and is usually divided into a first and second
carbonatation. There can remain enough antifoam within the sugar juice to
be active at this stage. However, it is common to add antifoams into the
lime milk tank (up to 5 g/t of beets), sometimes into the preliming vessel
(up to 10 g/t) and rarely into the first carbonatation.
Surfactant Antlfoams 317

The juice is then filtered and sent to evaporators to obtain a thick juice
containing about 65% dissolved solids. Evaporation to a supersaturated sugar
solution is performed in vacuum pans whereupon crystallization is induced
by seeding. Small amounts of antifoams are used in the vacuum pans. These
products will also reduce the viscosity and the surface tension and improve
the crystal growth. Several centrifuge stages are employed to extract the
maximum sugar. The remaining liquid, called molasses, can be used to pro-
duce alcohol by fermentation or sold to animal feed, yeast, citric acid, or
amino acid manufacturers.

C. Causes of Foam in the Sugar Beet Industry


1. Transport and Washing Stage
At the transport/washing stage damage to the skin of the sugar beet root
often occurs. This leads to release of surface-active compounds like saponins
which are primarily located just below the skin of the root. Foam problems
which occur at this stage may be in part attributed to such compounds. They
are also exacerbated by the presence of surface active vegetable components
derived from the soil which adhere in significant quantities to the beets (see
Table 6). In rainy periods, soil contamination is greater, the concentration
of vegetable impurities is higher and foaming more severe. This leads to
higher antifoam demand.

2. Diffusion and Postdiffusion


a. Components of Beet Liquors. There exists a tremendous number of
organic compounds in the sugar juice [29]. In addition to sucrose, by far
the greatest one, the following components have been identified:
Nitrogen-free substances.
Other carbohydrates in small amounts, mostly glucose, fructose, rafinose,
and kestoses
Traces of arabinose and gelectose
Various mono-, di-, and tricarboxylic acids, both unsubstituted and hydroxy
acids
Small quantities of pectic material
Saponins
Traces of numerous other organic compounds (vegetable fats made up of
lecithin, oleic, palmitic, and erucic acids)
Nitrogen compounds. As illustrated in Table 7, protein fragments are the
largest fraction and betaine is the largest individual component.
It has often been mentioned that saponine, a glycoside of oleanolic acid,
is a prime cause of foaming. Some tests [35) showed that the addition of
318 Blease et al.

TABLE 7 Nitrogenous Materials in Beets

% %
on beets of total

Amino acids and others 0.042 21.0


Amides 0.015 7.5
Ammonia 0.005 2.5
Nitrates 0.002 1.0
Proteins and peptides 0.115 27.5
Betane and choline 0.020 10.0
Purines 0.001 0.5
Total 0.200 100.0
Source: Ref. 33 and 34.

saponine and/or oleanolic acid to a normal production sugar did increase


the foaming tendency. It was however, concluded that the main effect was
due to the presence of peptides (amphoteric surface-active agents). Proteins
of the sugar juice are strongly surface active and give rise to foam [34].
b. Concentrations. The saponin contents of the beet peeling and the crown
were found to be 570 and 120 mg/kg, respectively [36]. In a diffusion juice,
it could reach up to 3673 mg/kg of dry substance [37], thus about 580 mg/
L of sugar juice. Some papers [37] even mention higher levels, up to 9193
mg/kg of dry substance.
c. Surface Tension. It was found f38] that the surface tension of saponin
solutions dropped very quickly down to 42 mN/m when the concentration
reached 25 mg/L. Another study [39] mentions that the surface tension of
diluted molasses could decrease down to 39 mN/m. The surface tension of
liquors from sugar beets harvested 1 hr before the test can vary between 60
and 53 mN/m [40] from the beginning to the end of the campaign. In ad-
dition, it was also shown that the surface tension can significantly decrease
when the storage time of the beets does increase. Both factors may explain
why the foaming difficulties at end of campaigns are usually more stringent
than at the beginning.

D. Composition of the Antifoam


As indicated above, sugar beet antifoams are proprietary mixtures, usually
complex blends, and the exact composition is not generally known.
1. Transport and Washing of Beets
Water temperatures during the beet harvest season are low (usually 5-l5°C).
Copolymers will only function effectively as antifoam ingredients at tern-
Surfactant Antifoams 319

TABLE 8 Typical Composition of a Beet-Washing Antifoam

Before 1975 Nowadays

Mineral oil 50% 50%


Animal and/or vegetable acid oils 50% 15-25%
Copolymer 15-25%
Alcohol 5-10%
Source: Ref. 17.

peratures above their respective cloud points. At temperatures below their


cloud points they are too water soluble and may even exacerbate a foam
problem. Therefore copolymers used for transport/washing of beets should
have low cloud-point temperatures. Esterification of copolymers with cheap
fatty acid will reduce the cloud points if the copolymer itself is unacceptable
in this respect.
In the past, the antifoam agents used for the beet washing did not contain
copolymers components. Since 1975 such polyalkyleneglycol copolymers
have become important ingredients, representing some 15%-25% of oil-
based formulations. The improvement in antifoam efficiency has allowed
consumption for this stage to decrease down to 20-60 g/tonne of beet, halv-
ing the previous dosage level.
The few patents which deal with sugar antifoams usually do not refer to
beet washing. However a typical composition is shown in Table 8.
The alcohol employed is usually isopropanol or isobutanol permitting a
homogeneous composition with a low pourpoint to be obtained. The anti-
foam added into the supply water of the water guns or in the beet washers
must have an immediate action. The high pressures/flowrates at these points
will pennit the recipes given in Table 8 to disperse. However, the antifoam
added in the clean water tank must have a preventative action. It must dis-
perse in water in order to be carried and be active at the critical points
downstream. Thus the formulator may add up to 5% of a nonionic surfactant
to the above-mentioned recipe.
2. Sugar Extraction: The Diffusion
The diffuser is also a critical part of the sugar extraction process as far as
foam is concerned due to the natural foaming tendency of the sugar juice
and due to the high volume and flowrates involved in some diffusers. In the
patent literature several types of surfactants are mentioned as being used in
the sugar diffuser antifoams.
In the United States there are only a limited number of surfactants allowed
by the Food and Drug Administration [41] to be used in sugar antifoam
320 Blease et al.

TABLE 9 Typical Composition of a Diffusion Antifoam

Before 1975 Nowadays


Acid fish oil 70%
Acid vegetable oil 20%
Fatty acid 30%
Mineral oil 10%
Copolymer 70%
Source: Ref. 17.

recipes. These are restricted to

A n-bµtoxy polyoxyethylene/polyoxypropylene copolymer


Polyoxyethylene (400) or (600) dioleate
Polyoxyethylene (600) monoricinoleate
Polypropylene glycol (MW 1200 to 2500)
Propylene glycol mono- and diesters of fats and fatty acids
Polysorbate 80 (polyoxyethylene (20) sorbitan monooleate).

As a consequence one patent [42] refers to compositions such as 40%-60%


sugar cane oil + 5%-10% emulsifier + 30%-50% mineral oil, a preferred
emulsifier being polyethylene glycol 400 monooleate. Another patent [30]
covers blends such as 2%-5% ethylenebis(stearamide) + 5%-10% poly-
ethyleneglycol (MW 400 to 600) and up to 15% fatty alcohol ethoxylate
(C11-1s + 7 EO) in mineral oil.
In other countries, blends containing up to 80% of alcohol alkoxylates,
like C 12; 18 + 5 EO + 13 PO, have been suggested [43]. Another patent [25]
covers the following composition 65% of a polypropyleneglycol (MW 2000)
+ 15% of polyethyleneglycol dioleate + 10% of a polypropyleneglycol
monooleate + 8% isopropanol + 2% polyethoxylated sorbitan monooleate.
Esters of fatty acids with polyoxyethylene/polyoxypropylene copolymers
are also mentioned [44), and the use of esters of nonylphenol ethoxylates
with fatty acids is claimed.
Esters of fatty acids with polyoxyethylene/polyoxypropylene copolymers
initiated on ethylene diamine would also be effective [45]. Reaction products
of a carbohydrate, a fatty acid, and an alkylene oxide [46] have also been
claimed. More recently [47], the use of an alkyl end-capped alcohol alkox-
ylate has been patented. A typical composition of a diffusion antifoam is
given in Table 9. The "old" recipe (used up to 1975) had to be heated to
be handled, but this is no longer the case with the formulation used today.
Surfactant Antifoams 321

3. Other Stages
a. Carbonatation. Antifoams are sometimes used during the preparation
of the milk or lime, during the preliming, and rarely during the carbona-
tation. In most cases it is common to use the same antifoam as the one used
at the diffusion stage.
b. Vacuum Pans. Small amounts, less than 5 g of a product per tonne
of massecuite (the concentrated sugar syrup) are added before applying vac-
uum. This product will avoid the formation of large air bubbles which would
burst at the surface and generate some losses. Some of them are claimed to
reduce the viscosity and surface tension of the massecuite and will generate
improved crystal growth. The products [48,49] which are used are coconut
oil, methyglucoside coconut oil ester, fatty alcohol alkoxylates, etc.
c. Fermentation. Some plants carry out a fermentation of the molasses
on the sugar juice with yeasts to produce alcohol. A medium plant will
produce up to 100,000 L of alcohol/day and a large one up to 200,000 L/
day. Copolymers blended with vegetable fatty acids are also largely used
as antifoams. The amounts of antifoams used during the fermentation will
be